{"as_of":"2026-08-16T02:39:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:69d0ab4d968702c1b36c9a68c5332bee37204a8abaaeda4f9db9c3a2c28282af","coverage":[{"denominator":0,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":36,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":36,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-15T06:32:42.880941+00:00","state":"measured"},{"denominator":36,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":36,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-15T22:16:38.293770Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":1,"source":"pith","source_observed_at":"2026-08-05T02:28:24.338817Z","state":"measured"}],"external_citation_measurements":[{"count":1,"observed_at":"2026-08-05T02:28:24.338817Z","source":"pith"}],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":"2503.01141","doi":"10.48550/arxiv.2503.01141","metadata_source":"pith","pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach","venue":"cs.CL","work_id":"21e3a5ff-8dd1-4b00-823c-baeba601afb5","year":2025},"citing_paper":{"arxiv_id":"2503.16419","last_updated":"2025-08-21T19:14:40Z","snapshot_observed_at":"2026-08-11T13:10:23.709172Z","submitted_at":"2025-03-20T17:59:38Z","title":"Stop Overthinking: A Survey on Efficient Reasoning for Large Language Models","version":4},"reference_index":83,"source":"pdf_text","source_observed_at":"2026-05-14T01:29:56.480020Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2503.16419"},"observation_digest":"sha256:3e025ddea38f07b5331b99fa13bca4bd6419a330c255f5674ccd21ee93414fe0","observation_id":"4ab46e37-6808-4078-adcb-06aeeede12c2","resolution":{"observed_at":"2026-05-14T01:29:56.978183Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-07-18T08:21:02.583941+00:00","source":"crossref_status_cache"},{"observed_at":"2026-07-18T08:21:02.583941+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-15T22:16:38.293770Z","title":"How well do llms compress their own chain-of- thought? a token complexity approach","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2505.07686","last_updated":"2025-05-17T04:01:57Z","snapshot_observed_at":"2026-08-15T22:08:07.921829Z","submitted_at":"2025-05-12T15:50:44Z","title":"S-GRPO: Early Exit via Reinforcement Learning in Reasoning Models","version":2},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-15T22:16:38.293770Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2505.07686"},"observation_digest":"sha256:f673a6b5e75a97776bd9aa077ea153d0d814aada15e74d2c7c0d35de2350bb04","observation_id":"1bb5f17b-3019-468e-82c1-e8061d248a69","resolution":{"observed_at":"2026-08-15T22:16:38.293770Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-15T20:52:27.824350Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach.arXiv preprint arXiv:2503.01141, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2505.11827","last_updated":"2025-05-25T04:09:29Z","snapshot_observed_at":"2026-08-15T20:44:40.659527Z","submitted_at":"2025-05-17T04:26:39Z","title":"Not All Thoughts are Generated Equal: Efficient LLM Reasoning via Multi-Turn Reinforcement Learning","version":2},"reference_index":55,"source":"pdf_text","source_observed_at":"2026-08-15T20:52:27.824350Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2505.11827"},"observation_digest":"sha256:a8844ac2f8d41f3ce72a9ede08b57cc650d9a3e0fc9265c7ba10735d9e2edcb7","observation_id":"776fd629-9eb6-44f0-b51e-a5c6e1992dcb","resolution":{"observed_at":"2026-08-15T20:52:27.824350Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-15T20:51:49.297164Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach.arXiv preprint arXiv:2503.01141, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2505.11896","last_updated":"2025-05-25T13:39:29Z","snapshot_observed_at":"2026-08-15T20:43:10.391637Z","submitted_at":"2025-05-17T08:27:00Z","title":"AdaCoT: Pareto-Optimal Adaptive Chain-of-Thought Triggering via Reinforcement Learning","version":2},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-15T20:51:49.297164Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2505.11896"},"observation_digest":"sha256:d7f4e394a5fb2d06f9b30e85742951499af0730a1511dcd9329ed591807138d7","observation_id":"3c2359e8-6bf5-4eaf-ad61-da81ef55338c","resolution":{"observed_at":"2026-08-15T20:51:49.297164Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-07T15:29:14.562977Z","title":"How well do llms compress their own chain-of- thought? a token complexity approach","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2505.15154","last_updated":"2025-05-21T06:20:17Z","snapshot_observed_at":"2026-08-16T02:15:45.088243Z","submitted_at":"2025-05-21T06:20:17Z","title":"Prolonged Reasoning Is Not All You Need: Certainty-Based Adaptive Routing for Efficient LLM/MLLM Reasoning","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-07T15:29:14.562977Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2505.15154"},"observation_digest":"sha256:787ef71f468037dfa63160725a8a03e975754e817cf258f1dda609a8ef126c0c","observation_id":"5cbcb441-eacb-47ef-9162-40e8d2112cc9","resolution":{"observed_at":"2026-08-07T15:29:14.562977Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-07T14:45:12.973380Z","title":null,"venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2505.17827","last_updated":"2025-08-02T10:54:47Z","snapshot_observed_at":"2026-08-15T20:25:22.727186Z","submitted_at":"2025-05-23T12:41:29Z","title":"Not All Tokens Are What You Need In Thinking","version":2},"reference_index":11,"source":"arxiv_source","source_observed_at":"2026-08-07T14:45:12.973380Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2505.17827"},"observation_digest":"sha256:9dfc43a00c17a7dc72e263ee9b14b2ac618d8ef95c4155c8933026daa59161d1","observation_id":"8eeefb32-8546-441d-a421-fa6719881067","resolution":{"observed_at":"2026-08-07T14:45:12.973380Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-07T14:42:12.178937Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2505.17941","last_updated":"2025-05-23T14:17:56Z","snapshot_observed_at":"2026-08-15T23:32:49.904225Z","submitted_at":"2025-05-23T14:17:56Z","title":"VeriThinker: Learning to Verify Makes Reasoning Model Efficient","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-07T14:42:12.178937Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2505.17941"},"observation_digest":"sha256:be5da4267fff27f0978454bbd3eb8fac86245ec96e646599ee806fa618e17145","observation_id":"8987be17-15f5-4a5f-b16b-26af886cf3ac","resolution":{"observed_at":"2026-08-07T14:42:12.178937Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-07T14:27:38.910871Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2505.18962","last_updated":"2025-05-31T15:08:41Z","snapshot_observed_at":"2026-08-13T11:40:13.393602Z","submitted_at":"2025-05-25T03:35:49Z","title":"System-1.5 Reasoning: Traversal in Language and Latent Spaces with Dynamic Shortcuts","version":3},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-07T14:27:38.910871Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2505.18962"},"observation_digest":"sha256:860012866985041be679647ac471d61804c7819f75f6e3f44aa20b11889748e1","observation_id":"30336f93-f526-486f-a932-954128d5133c","resolution":{"observed_at":"2026-08-07T14:27:38.910871Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-07T13:53:56.890817Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach.arXiv preprint arXiv:2503.01141,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2505.20643","last_updated":"2025-05-27T02:44:00Z","snapshot_observed_at":"2026-08-14T22:00:25.911573Z","submitted_at":"2025-05-27T02:44:00Z","title":"Can Past Experience Accelerate LLM Reasoning?","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-07T13:53:56.890817Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2505.20643"},"observation_digest":"sha256:f37f6ed8385b6db1dbcc5ac9a362aa817ec1d5836b0fb7ff0f0f72b2ba187fbf","observation_id":"403a5b76-5ee4-4c61-a5f3-96edc1b40eb2","resolution":{"observed_at":"2026-08-07T13:53:56.890817Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-07T10:52:43.571421Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2506.04182","last_updated":"2025-06-04T17:28:38Z","snapshot_observed_at":"2026-08-13T02:34:18.465840Z","submitted_at":"2025-06-04T17:28:38Z","title":"Long or short CoT? Investigating Instance-level Switch of Large Reasoning Models","version":1},"reference_index":13,"source":"arxiv_source","source_observed_at":"2026-08-07T10:52:43.571421Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2506.04182"},"observation_digest":"sha256:0ea27bc7420eddef28e7e74ff6282b3b3e1c4884984e133e4a3c32bade3a462c","observation_id":"24d68348-f322-4e3b-af59-53b5ea542bf5","resolution":{"observed_at":"2026-08-07T10:52:43.571421Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-07T05:21:03.148000Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach,","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2506.08422","last_updated":"2025-06-11T03:16:55Z","snapshot_observed_at":"2026-08-15T21:13:08.414357Z","submitted_at":"2025-06-10T03:48:26Z","title":"Transforming Expert Knowledge into Scalable Ontology via Large Language Models","version":2},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-07T05:21:03.148000Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2506.08422"},"observation_digest":"sha256:be203c4610ac2e22e3c3f1366936b0e0077879ba4fa6a952358832a81dc13820","observation_id":"075382ea-dd8f-4d57-8f07-0c324621f3d4","resolution":{"observed_at":"2026-08-07T05:21:03.148000Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-07T04:38:09.478670Z","title":null,"venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2506.10446","last_updated":"2025-06-12T07:49:24Z","snapshot_observed_at":"2026-08-13T09:20:49.647528Z","submitted_at":"2025-06-12T07:49:24Z","title":"Fast on the Easy, Deep on the Hard: Efficient Reasoning via Powered Length Penalty","version":1},"reference_index":15,"source":"arxiv_source","source_observed_at":"2026-08-07T04:38:09.478670Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2506.10446"},"observation_digest":"sha256:35fba960f5e4f331146f1bf08a51eb11a8ee4076b6f554b6d96452610a2fe9d5","observation_id":"1effa212-a6a7-4461-8958-7e997de45ed1","resolution":{"observed_at":"2026-08-07T04:38:09.478670Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-07T04:26:29.051959Z","title":null,"venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2506.10716","last_updated":"2025-06-12T14:05:09Z","snapshot_observed_at":"2026-08-15T16:21:04.205867Z","submitted_at":"2025-06-12T14:05:09Z","title":"PREMISE: Scalable and Strategic Prompt Optimization for Efficient Mathematical Reasoning in Large Models","version":1},"reference_index":17,"source":"arxiv_source","source_observed_at":"2026-08-07T04:26:29.051959Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2506.10716"},"observation_digest":"sha256:b8acf6c42f01b249ed680e23b2f372e8c74ecd9d7b9ec304077c08c66e0a4143","observation_id":"c7667202-731f-4a95-8a9c-6945d5bf36b0","resolution":{"observed_at":"2026-08-07T04:26:29.051959Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-07T04:24:02.269001Z","title":null,"venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2506.10822","last_updated":"2025-06-12T15:43:01Z","snapshot_observed_at":"2026-08-15T19:41:26.570388Z","submitted_at":"2025-06-12T15:43:01Z","title":"ReCUT: Balancing Reasoning Length and Accuracy in LLMs via Stepwise Trails and Preference Optimization","version":1},"reference_index":15,"source":"arxiv_source","source_observed_at":"2026-08-07T04:24:02.269001Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2506.10822"},"observation_digest":"sha256:fd9ed7d4a0e81f9787092034a28873b5e2ccdffa6f7537d88c346b7dd48b0d0e","observation_id":"ba57bb8a-0c5c-4241-93b9-14de7bab1e93","resolution":{"observed_at":"2026-08-07T04:24:02.269001Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-07T00:57:01.431811Z","title":"How well do Ilms compress their own chain-of- thought? a token complexity approach","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.12353","last_updated":"2025-06-14T05:30:09Z","snapshot_observed_at":"2026-08-07T08:59:45.729814Z","submitted_at":"2025-06-14T05:30:09Z","title":"Efficient Reasoning Through Suppression of Self-Affirmation Reflections in Large Reasoning Models","version":1},"reference_index":2023,"source":"pdf_text","source_observed_at":"2026-08-07T00:57:01.431811Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2506.12353"},"observation_digest":"sha256:dfef2afff064a3f70d8751fc611ef68f6afea1bf7cdc450c932d9ed7bd410cc4","observation_id":"c508e0f9-1bb8-4906-9b2b-9da57a366953","resolution":{"observed_at":"2026-08-07T00:57:01.431811Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-06T23:10:25.880397Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2506.19492","last_updated":"2025-06-24T10:25:28Z","snapshot_observed_at":"2026-08-14T11:12:47.256853Z","submitted_at":"2025-06-24T10:25:28Z","title":"Is Long-to-Short a Free Lunch? Investigating Inconsistency and Reasoning Efficiency in LRMs","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-06T23:10:25.880397Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2506.19492"},"observation_digest":"sha256:e51e318e5c74f692bfb4bbcb4d052372ec548265503b97d87c7b67e9fd11a3a4","observation_id":"91f29c4a-bcc5-4e15-8794-2a82d943533f","resolution":{"observed_at":"2026-08-06T23:10:25.880397Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-07T05:07:39.996793Z","title":"How well do LLMs compress their own chain-of-thought? a token complexity approach,","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2507.00004","last_updated":"2025-07-10T17:08:40Z","snapshot_observed_at":"2026-08-12T15:42:28.437403Z","submitted_at":"2025-06-10T14:47:48Z","title":"A Theory of Inference Compute Scaling: Reasoning through Directed Stochastic Skill Search","version":2},"reference_index":122,"source":"pdf_text","source_observed_at":"2026-08-07T05:07:39.996793Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2507.00004"},"observation_digest":"sha256:0b3881f352373310570a611b9ab1f84a3274d8a34360dda5194946384f979a39","observation_id":"aa3b00cc-ddab-4cda-8919-57662c1559f5","resolution":{"observed_at":"2026-08-07T05:07:39.996793Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-06T20:43:09.527822Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach.arXiv preprint arXiv:2503.01141,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2507.02076","last_updated":"2025-07-02T18:27:42Z","snapshot_observed_at":"2026-08-14T10:43:51.935390Z","submitted_at":"2025-07-02T18:27:42Z","title":"Reasoning on a Budget: A Survey of Adaptive and Controllable Test-Time Compute in LLMs","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-06T20:43:09.527822Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2507.02076"},"observation_digest":"sha256:96123d13312704395c26594d4e796541fffca43980228fdd9fe930c2334512f3","observation_id":"5c0c275e-816f-4fe9-92d5-9b3b91d06193","resolution":{"observed_at":"2026-08-06T20:43:09.527822Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-06T19:16:31.318760Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2507.06087","last_updated":"2025-07-08T15:28:48Z","snapshot_observed_at":"2026-08-11T12:26:13.493802Z","submitted_at":"2025-07-08T15:28:48Z","title":"CoRE: Enhancing Metacognition with Label-free Self-evaluation in LRMs","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-06T19:16:31.318760Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2507.06087"},"observation_digest":"sha256:d87fd8446448e1590768a560536eac552dfc2023a5a72109cd508e1c71c50ceb","observation_id":"b1cff093-9b88-4da9-be5c-07ed63605c80","resolution":{"observed_at":"2026-08-06T19:16:31.318760Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-06T17:54:16.926500Z","title":null,"venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2507.09662","last_updated":"2025-07-13T14:51:59Z","snapshot_observed_at":"2026-08-15T08:55:38.744613Z","submitted_at":"2025-07-13T14:51:59Z","title":"Towards Concise and Adaptive Thinking in Large Reasoning Models: A Survey","version":1},"reference_index":92,"source":"arxiv_source","source_observed_at":"2026-08-06T17:54:16.926500Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2507.09662"},"observation_digest":"sha256:b59c18c82861096c8ae1af7d33a52e3e22edc6d94921d82d0220fbfd3a7573b0","observation_id":"eeb2d815-694a-47a1-afe4-d495821f0d78","resolution":{"observed_at":"2026-08-06T17:54:16.926500Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-06T15:20:13.782725Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2507.16331","last_updated":"2026-07-03T09:17:03Z","snapshot_observed_at":"2026-08-15T12:22:31.317936Z","submitted_at":"2025-07-22T08:13:01Z","title":"Re:Form -- Reducing Human Annotations in Scalable Formal Software Verification with RL in LLMs: A Preliminary Study on Dafny","version":4},"reference_index":39,"source":"arxiv_source","source_observed_at":"2026-08-06T15:20:13.782725Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2507.16331"},"observation_digest":"sha256:53ec9a98f5447a492f0de4293addfa28546679e77141f3bd2fdbe1824be4c1d4","observation_id":"574e74cd-6a1e-41ad-b3c1-a22270773492","resolution":{"observed_at":"2026-08-06T15:20:13.782725Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-05T12:49:08.451125Z","title":null,"venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2509.01236","last_updated":"2025-09-01T08:24:28Z","snapshot_observed_at":"2026-08-14T21:12:33.893597Z","submitted_at":"2025-09-01T08:24:28Z","title":"Rethinking the Chain-of-Thought: The Roles of In-Context Learning and Pre-trained Priors","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-05T12:49:08.451125Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2509.01236"},"observation_digest":"sha256:1fdf475b0621bca1379a3c05923bd6ef6184ceff62fa8deb1865e878a17f7f8d","observation_id":"961f2d5d-c8f6-4927-a288-4176ebef7c33","resolution":{"observed_at":"2026-08-05T12:49:08.451125Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-04T07:59:52.911656Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach.arXiv preprint arXiv:2503.01141,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2510.23486","last_updated":"2026-07-03T18:03:08Z","snapshot_observed_at":"2026-08-09T00:53:49.879051Z","submitted_at":"2025-10-27T16:17:45Z","title":"Learning to Reason Efficiently with Discounted Reinforcement Learning","version":3},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-04T07:59:52.911656Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2510.23486"},"observation_digest":"sha256:950f11fed08b774ee6198cb7a9cb705d23192bcf8bf77f85274258a3312901e4","observation_id":"53ffe2a6-1c30-4b63-99f7-f68e480c8e67","resolution":{"observed_at":"2026-08-04T07:59:52.911656Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-03T05:17:30.015340Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2602.02823","last_updated":"2026-05-29T19:48:02Z","snapshot_observed_at":"2026-08-03T05:17:28.452364Z","submitted_at":"2026-02-02T21:23:51Z","title":"R2-Router: A New Paradigm for LLM Routing with Reasoning","version":2},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-03T05:17:30.015340Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2602.02823"},"observation_digest":"sha256:6a8c55b61e0939c6dc386032a1468861d9f43fadcac43a07829f629be8fb3723","observation_id":"f096145a-9803-44a8-b391-cb1cb44945ab","resolution":{"observed_at":"2026-08-03T05:17:30.015340Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-03T05:19:45.889513Z","title":"Lee, A., Che, E., and Peng, T","venue":null,"work_id":null,"year":1965},"citing_paper":{"arxiv_id":"2602.02909","last_updated":"2026-05-27T22:56:08Z","snapshot_observed_at":"2026-08-11T15:00:53.761740Z","submitted_at":"2026-02-02T23:33:34Z","title":"Reasoning about Reasoning: BAPO Bounds on Chain-of-Thought Token Complexity in LLMs","version":2},"reference_index":1965,"source":"pdf_text","source_observed_at":"2026-08-03T05:19:45.889513Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2602.02909"},"observation_digest":"sha256:ead17d29e405058c04316d36fdab3c5474677e73f3736f1780c5d801e44e54e7","observation_id":"31776a65-cbc2-4a41-bdf3-4db73b6d2c6d","resolution":{"observed_at":"2026-08-03T05:19:45.889513Z","resolver_source":null,"status":"malformed_identifier"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":"2503.01141","doi":"10.48550/arxiv.2503.01141","metadata_source":"pith","pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach","venue":"cs.CL","work_id":"21e3a5ff-8dd1-4b00-823c-baeba601afb5","year":2025},"citing_paper":{"arxiv_id":"2602.08324","last_updated":"2026-05-25T15:41:17Z","snapshot_observed_at":"2026-08-16T00:54:27.830552Z","submitted_at":"2026-02-09T06:57:15Z","title":"Towards Efficient Large Language Reasoning Models via Extreme-Ratio Chain-of-Thought Compression","version":3},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-05-21T13:43:51.127429Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2602.08324"},"observation_digest":"sha256:7e4beead3c49c1ea5b6f341e65a26373fdd1f21ef33d3158e4b6d34d6af922c3","observation_id":"b16a5ea6-cba5-4782-8300-d858b58c058a","resolution":{"observed_at":"2026-05-21T13:44:11.436773Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-07-18T08:21:02.583941+00:00","source":"crossref_status_cache"},{"observed_at":"2026-07-18T08:21:02.583941+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-03T03:25:23.079397Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2602.08324","last_updated":"2026-05-25T15:41:17Z","snapshot_observed_at":"2026-08-16T00:54:27.830552Z","submitted_at":"2026-02-09T06:57:15Z","title":"Towards Efficient Large Language Reasoning Models via Extreme-Ratio Chain-of-Thought Compression","version":4},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-03T03:25:23.079397Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2602.08324"},"observation_digest":"sha256:38981a9a756c5a1f04522af00a186eef08dd2ee1b45995cf72a231139be4f22f","observation_id":"078b5a79-d31b-40a2-91fc-5d0ebe7a4ad1","resolution":{"observed_at":"2026-08-03T03:25:23.079397Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":"2503.01141","doi":"10.48550/arxiv.2503.01141","metadata_source":"pith","pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach","venue":"cs.CL","work_id":"21e3a5ff-8dd1-4b00-823c-baeba601afb5","year":2025},"citing_paper":{"arxiv_id":"2604.17002","last_updated":"2026-04-18T14:18:13Z","snapshot_observed_at":"2026-08-11T03:04:11.475105Z","submitted_at":"2026-04-18T14:18:13Z","title":"Intelligent Drill-Down: Large Language Model-Driven Drill-Down Technique for Human-AI Collaborative Visual Exploration","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-05-10T06:37:21.309366Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2604.17002"},"observation_digest":"sha256:19c3445f51ea9f075d1c8494f7cbb6e7f86406aae33c31d27b836c98ac58c2df","observation_id":"ad88d40c-23df-4e43-8706-e7f47dca5537","resolution":{"observed_at":"2026-05-10T06:41:36.861051Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-07-18T08:21:02.583941+00:00","source":"crossref_status_cache"},{"observed_at":"2026-07-18T08:21:02.583941+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":"2503.01141","doi":"10.48550/arxiv.2503.01141","metadata_source":"pith","pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach","venue":"cs.CL","work_id":"21e3a5ff-8dd1-4b00-823c-baeba601afb5","year":2025},"citing_paper":{"arxiv_id":"2605.06165","last_updated":"2026-05-07T12:51:49Z","snapshot_observed_at":"2026-07-06T23:18:41.400741Z","submitted_at":"2026-05-07T12:51:49Z","title":"Post Reasoning: Improving the Performance of Non-Thinking Models at No Cost","version":1},"reference_index":109,"source":"arxiv_source","source_observed_at":"2026-05-08T10:19:08.451445Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2605.06165"},"observation_digest":"sha256:ca352d1fb09c32c48734985c226890940871b86008c30d803c8d29ced267c452","observation_id":"0e63e27d-0aa5-402b-af25-2f9c0a12ceb5","resolution":{"observed_at":"2026-05-11T20:06:09.377376Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-07-18T08:21:02.583941+00:00","source":"crossref_status_cache"},{"observed_at":"2026-07-18T08:21:02.583941+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":"2503.01141","doi":"10.48550/arxiv.2503.01141","metadata_source":"pith","pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach","venue":"cs.CL","work_id":"21e3a5ff-8dd1-4b00-823c-baeba601afb5","year":2025},"citing_paper":{"arxiv_id":"2605.08776","last_updated":"2026-05-09T08:04:47Z","snapshot_observed_at":"2026-08-15T18:08:48.298094Z","submitted_at":"2026-05-09T08:04:47Z","title":"Reasoning Compression with Mixed-Policy Distillation","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-05-12T01:29:02.387691Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2605.08776"},"observation_digest":"sha256:bf91cf43113b3722f6847bc8819c0b70317afdaa033f843708881d3a5e3a2ba7","observation_id":"f6727f4c-34e4-401c-ab16-6ebfaeb57273","resolution":{"observed_at":"2026-05-12T07:56:28.234223Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-07-18T08:21:02.583941+00:00","source":"crossref_status_cache"},{"observed_at":"2026-07-18T08:21:02.583941+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":"2503.01141","doi":"10.48550/arxiv.2503.01141","metadata_source":"pith","pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach","venue":"cs.CL","work_id":"21e3a5ff-8dd1-4b00-823c-baeba601afb5","year":2025},"citing_paper":{"arxiv_id":"2605.30832","last_updated":"2026-05-29T04:37:49Z","snapshot_observed_at":"2026-07-06T23:40:03.151978Z","submitted_at":"2026-05-29T04:37:49Z","title":"SLAT: Segment-Level Adaptive Trimming for Efficient CoT Reasoning","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-06-28T22:27:30.783923Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2605.30832"},"observation_digest":"sha256:74b8d2552d9a1cafa521b6dc30c6b1fb895c8663a63605d24ca98cb49f530b61","observation_id":"d8dd3893-56d5-4d31-bb15-a86debe7bb4a","resolution":{"observed_at":"2026-06-28T22:32:44.436704Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-07-18T08:21:02.583941+00:00","source":"crossref_status_cache"},{"observed_at":"2026-07-18T08:21:02.583941+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":"2503.01141","doi":"10.48550/arxiv.2503.01141","metadata_source":"pith","pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach","venue":"cs.CL","work_id":"21e3a5ff-8dd1-4b00-823c-baeba601afb5","year":2025},"citing_paper":{"arxiv_id":"2606.01168","last_updated":"2026-05-31T11:20:00Z","snapshot_observed_at":"2026-08-15T12:13:34.288229Z","submitted_at":"2026-05-31T11:20:00Z","title":"Thinking Economically: A Hierarchical Framework for Adaptive-Complexity Reasoning in LLMs","version":1},"reference_index":53,"source":"arxiv_source","source_observed_at":"2026-06-28T17:05:48.244094Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2606.01168"},"observation_digest":"sha256:567606d01419c5929aae5ebfee59af2a5d7c27fa013ad22eeb16a03852eddb1d","observation_id":"963b4978-f4aa-401a-94ce-a46907a1bd5b","resolution":{"observed_at":"2026-06-28T17:12:25.244358Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-07-18T08:21:02.583941+00:00","source":"crossref_status_cache"},{"observed_at":"2026-07-18T08:21:02.583941+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":"2503.01141","doi":"10.48550/arxiv.2503.01141","metadata_source":"pith","pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach","venue":"cs.CL","work_id":"21e3a5ff-8dd1-4b00-823c-baeba601afb5","year":2025},"citing_paper":{"arxiv_id":"2606.03503","last_updated":"2026-06-02T11:21:27Z","snapshot_observed_at":"2026-08-14T15:31:35.794905Z","submitted_at":"2026-06-02T11:21:27Z","title":"ThoughtFold: Folding Reasoning Chains via Introspective Preference Learning","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-06-28T10:07:16.700499Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2606.03503"},"observation_digest":"sha256:72fe70c798a72f6050b48169e5a037754b47fdfbed478b8a4a49ee30baa468d3","observation_id":"26cf7ce9-43bb-4cb4-a279-853901432f8d","resolution":{"observed_at":"2026-07-02T03:26:28.750825Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-07-18T08:21:02.583941+00:00","source":"crossref_status_cache"},{"observed_at":"2026-07-18T08:21:02.583941+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":"2503.01141","doi":"10.48550/arxiv.2503.01141","metadata_source":"pith","pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach","venue":"cs.CL","work_id":"21e3a5ff-8dd1-4b00-823c-baeba601afb5","year":2025},"citing_paper":{"arxiv_id":"2606.17890","last_updated":"2026-06-16T13:10:30Z","snapshot_observed_at":"2026-08-12T14:46:48.192265Z","submitted_at":"2026-06-16T13:10:30Z","title":"Dynamic Rollout Editing for Reducing Overthinking in RL-Trained Reasoning Models","version":1},"reference_index":23,"source":"arxiv_source","source_observed_at":"2026-06-27T01:19:55.164835Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2606.17890"},"observation_digest":"sha256:9c62f93d6033ed0f86e83e0fb7709f5de475ccbaa0fafd719ce9b69e05e26ef9","observation_id":"94a043c8-4983-4d23-a67b-97e9b9be399f","resolution":{"observed_at":"2026-06-27T01:20:20.433389Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-07-18T08:21:02.583941+00:00","source":"crossref_status_cache"},{"observed_at":"2026-07-18T08:21:02.583941+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":"2503.01141","doi":"10.48550/arxiv.2503.01141","metadata_source":"pith","pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"How well do llms compress their own chain-of-thought? a token complexity approach","venue":"cs.CL","work_id":"21e3a5ff-8dd1-4b00-823c-baeba601afb5","year":2025},"citing_paper":{"arxiv_id":"2607.08399","last_updated":"2026-07-09T12:21:44Z","snapshot_observed_at":"2026-08-05T01:38:41.823519Z","submitted_at":"2026-07-09T12:21:44Z","title":"Prompt Compression via Activation Aggregation","version":1},"reference_index":15,"source":"arxiv_source","source_observed_at":"2026-07-10T08:03:46.297577Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2607.08399"},"observation_digest":"sha256:0fd6aa340d4056bab6a32388705cbe35de91ec3156d2364a77a6eed3b27f6582","observation_id":"0d24e869-e3be-468e-a517-2a4a6f159c49","resolution":{"observed_at":"2026-07-10T08:06:57.494774Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-07-18T08:21:02.583941+00:00","source":"crossref_status_cache"},{"observed_at":"2026-07-18T08:21:02.583941+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.01141","snapshot_observed_at":"2026-08-01T18:49:28.691150Z","title":null,"venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.17188","last_updated":"2026-07-19T11:00:59Z","snapshot_observed_at":"2026-08-13T20:22:15.952538Z","submitted_at":"2026-07-19T11:00:59Z","title":"Is Your Model Thinking or Just Stagnating? PUMA: Diagnosing Reasoning Pathology via Phase-Momentum Alignment","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-01T18:49:28.691150Z"},"links":{"cited_paper":"/paper/2503.01141","citing_paper":"/paper/2607.17188"},"observation_digest":"sha256:167fe69d8510aa36d707fd1a5434a8e11f0587eff8749da0d1f97e7228c2ba66","observation_id":"e4ebe9b6-1058-4b5b-b329-185e335ac99a","resolution":{"observed_at":"2026-08-01T18:49:28.691150Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"links":{"evidence":"/evidence","html":"/paper/2503.01141/citation-record","integrity":"/paper/2503.01141/integrity","json":"/paper/2503.01141/citation-record.json","paper":"/paper/2503.01141"},"outbound":[],"paper":{"arxiv_id":"2503.01141","last_updated":"2025-04-01T00:41:36Z","latest_version":2,"primary_category":"cs.CL","snapshot_observed_at":"2026-08-08T04:02:43.828729Z","submitted_at":"2025-03-03T03:48:20Z","title":"How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach"},"reference_resolution":{"displayed":0,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":0,"verified_exact":0,"verified_fuzzy":0},"total_outbound_references":0},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"thesis":"As of 16 August 2026, this Paper Citation Record lists 0 of 0 outbound references and 36 inbound Pith citation observations for arXiv:2503.01141."}