{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:JDS6EKHCJRJKQAZVPG76T5FEAF","short_pith_number":"pith:JDS6EKHC","schema_version":"1.0","canonical_sha256":"48e5e228e24c52a8033579bfe9f4a4016b69ed5c2f3f8af865616f60611079c6","source":{"kind":"arxiv","id":"2608.02867","version":1},"attestation_state":"computed","paper":{"title":"BODHI: Do LLMs Branch Out and Discover Heterogeneous Inferences?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Akshay Chaturvedi, Krish Sharma, Nicholas Asher, Soumadeep Saha","submitted_at":"2026-08-03T20:37:54Z","abstract_excerpt":"Although reinforcement learning with verifiable rewards (RLVR) has improved the performance of large language models (LLMs) across a variety of reasoning tasks, there is significant debate as to whether RLVR expands the reasoning capability boundary, or just improves sampling efficiency. In this paper, we investigate the nature of test-time exploration in RLVR-trained LLMs by employing controlled maze-solving experiments and extracting a tree structure from mathematical reasoning traces (BODHI-Trees) based on semantic equivalence. This helps us delineate between entropy arising from stylistic "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2608.02867","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-08-03T20:37:54Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"7cd4ac21b07e43da3fc43fa7b0fe050c9ea873d513dbb2bfc79a64ff102ee7b0","abstract_canon_sha256":"d720ef08e2859282061724113172642bfd51b4a0c6615a83e8217b3024d03625"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-05T00:43:12.117340Z","signature_b64":"ZBxhPvqAA9UuMyiGng+XmJ5+XfCIFQiiD30vTictfVOWELLmPThfNMEkkwI64Udr1pt7PgpD5wq5EKTZ2tzPDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"48e5e228e24c52a8033579bfe9f4a4016b69ed5c2f3f8af865616f60611079c6","last_reissued_at":"2026-08-05T00:43:12.114735Z","signature_status":"signed_v1","first_computed_at":"2026-08-05T00:43:12.114735Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BODHI: Do LLMs Branch Out and Discover Heterogeneous Inferences?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Akshay Chaturvedi, Krish Sharma, Nicholas Asher, Soumadeep Saha","submitted_at":"2026-08-03T20:37:54Z","abstract_excerpt":"Although reinforcement learning with verifiable rewards (RLVR) has improved the performance of large language models (LLMs) across a variety of reasoning tasks, there is significant debate as to whether RLVR expands the reasoning capability boundary, or just improves sampling efficiency. In this paper, we investigate the nature of test-time exploration in RLVR-trained LLMs by employing controlled maze-solving experiments and extracting a tree structure from mathematical reasoning traces (BODHI-Trees) based on semantic equivalence. This helps us delineate between entropy arising from stylistic "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.02867","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2608.02867/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2608.02867","created_at":"2026-08-05T00:43:12.115721+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.02867v1","created_at":"2026-08-05T00:43:12.115721+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.02867","created_at":"2026-08-05T00:43:12.115721+00:00"},{"alias_kind":"pith_short_12","alias_value":"JDS6EKHCJRJK","created_at":"2026-08-05T00:43:12.115721+00:00"},{"alias_kind":"pith_short_16","alias_value":"JDS6EKHCJRJKQAZV","created_at":"2026-08-05T00:43:12.115721+00:00"},{"alias_kind":"pith_short_8","alias_value":"JDS6EKHC","created_at":"2026-08-05T00:43:12.115721+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JDS6EKHCJRJKQAZVPG76T5FEAF","json":"https://pith.science/pith/JDS6EKHCJRJKQAZVPG76T5FEAF.json","graph_json":"https://pith.science/api/pith-number/JDS6EKHCJRJKQAZVPG76T5FEAF/graph.json","events_json":"https://pith.science/api/pith-number/JDS6EKHCJRJKQAZVPG76T5FEAF/events.json","paper":"https://pith.science/paper/JDS6EKHC"},"agent_actions":{"view_html":"https://pith.science/pith/JDS6EKHCJRJKQAZVPG76T5FEAF","download_json":"https://pith.science/pith/JDS6EKHCJRJKQAZVPG76T5FEAF.json","view_paper":"https://pith.science/paper/JDS6EKHC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.02867&json=true","fetch_graph":"https://pith.science/api/pith-number/JDS6EKHCJRJKQAZVPG76T5FEAF/graph.json","fetch_events":"https://pith.science/api/pith-number/JDS6EKHCJRJKQAZVPG76T5FEAF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JDS6EKHCJRJKQAZVPG76T5FEAF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JDS6EKHCJRJKQAZVPG76T5FEAF/action/storage_attestation","attest_author":"https://pith.science/pith/JDS6EKHCJRJKQAZVPG76T5FEAF/action/author_attestation","sign_citation":"https://pith.science/pith/JDS6EKHCJRJKQAZVPG76T5FEAF/action/citation_signature","submit_replication":"https://pith.science/pith/JDS6EKHCJRJKQAZVPG76T5FEAF/action/replication_record"}},"created_at":"2026-08-05T00:43:12.115721+00:00","updated_at":"2026-08-05T00:43:12.115721+00:00"}