{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:KLGNQ6LALRFRY6T2HEKVRQOZ6U","short_pith_number":"pith:KLGNQ6LA","canonical_record":{"source":{"id":"2607.08572","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-09T15:04:05Z","cross_cats_sorted":[],"title_canon_sha256":"4414c5f9da24ca6125a4166338ad5dd24c02ba8be7fc88f8cd97af4a109ce738","abstract_canon_sha256":"b9d9cb76a8ff0814afaca653609be609b2632f32c9b65e84c0f0bfa2adaad3a9"},"schema_version":"1.0"},"canonical_sha256":"52ccd879605c4b1c7a7a391558c1d9f51eb7e641b3b5bbdfa266c7bc905d50cb","source":{"kind":"arxiv","id":"2607.08572","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.08572","created_at":"2026-07-10T01:19:53Z"},{"alias_kind":"arxiv_version","alias_value":"2607.08572v1","created_at":"2026-07-10T01:19:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.08572","created_at":"2026-07-10T01:19:53Z"},{"alias_kind":"pith_short_12","alias_value":"KLGNQ6LALRFR","created_at":"2026-07-10T01:19:53Z"},{"alias_kind":"pith_short_16","alias_value":"KLGNQ6LALRFRY6T2","created_at":"2026-07-10T01:19:53Z"},{"alias_kind":"pith_short_8","alias_value":"KLGNQ6LA","created_at":"2026-07-10T01:19:53Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:KLGNQ6LALRFRY6T2HEKVRQOZ6U","target":"record","payload":{"canonical_record":{"source":{"id":"2607.08572","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-09T15:04:05Z","cross_cats_sorted":[],"title_canon_sha256":"4414c5f9da24ca6125a4166338ad5dd24c02ba8be7fc88f8cd97af4a109ce738","abstract_canon_sha256":"b9d9cb76a8ff0814afaca653609be609b2632f32c9b65e84c0f0bfa2adaad3a9"},"schema_version":"1.0"},"canonical_sha256":"52ccd879605c4b1c7a7a391558c1d9f51eb7e641b3b5bbdfa266c7bc905d50cb","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-10T01:19:53.932932Z","signature_b64":"F0lcHmFoSsBaKhOYOzUFWllLcf7XQbtiA1xzZ5HXSK+Xk33JsI/yBo7kTEnrlNghcwoQ1ODq3mbLozUx22NlAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"52ccd879605c4b1c7a7a391558c1d9f51eb7e641b3b5bbdfa266c7bc905d50cb","last_reissued_at":"2026-07-10T01:19:53.932604Z","signature_status":"signed_v1","first_computed_at":"2026-07-10T01:19:53.932604Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.08572","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-10T01:19:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FoUkNOBdUCsZoyhMzQCC/DXd0c6LZET9itUJbiYXZAL2Dn0i2cpDfJKlwowcA+y4AJaZTryaatZjGkVxzDhqDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T10:55:30.534280Z"},"content_sha256":"191e0df71db655b1d9affb9bbb8ad964a148b4898d7bae5b9549cb4c36399264","schema_version":"1.0","event_id":"sha256:191e0df71db655b1d9affb9bbb8ad964a148b4898d7bae5b9549cb4c36399264"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:KLGNQ6LALRFRY6T2HEKVRQOZ6U","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Switch-Reasoner: Learn When to Think in Multitask Mixtures via Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jian Liang, Jian Luan, Jinjie Li, Mang Ye, Pei Fu, Ruijie Luo, Shaojie Zhang, Wenke Huang, Yi R. Fung, Yiyang Fang","submitted_at":"2026-07-09T15:04:05Z","abstract_excerpt":"Multimodal Large Language Models (MLLMs) often follow a fixed Think-then-Answer paradigm, which is inefficient in heterogeneous multitask settings because simple inputs may not require explicit reasoning while difficult ones can benefit substantially from it. Learning when to think is also unstable during post-training, where imbalanced rollouts can drive the model toward always-thinking or always-direct behavior. We propose Switch-Reasoner, a GRPO-based framework that learns to adaptively select reasoning modes for MLLMs. It treats thinking as a virtual tool invocation and allows the model to"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.08572","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/2607.08572/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-10T01:19:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AMb9M0gWlOnBtGVR1QFaHR6ZLwenzDP+gK3GLMH2+1Tyx21aUx67IxPuGBEh8GI7ChljUMnZpx5TrxxrdDphCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T10:55:30.534812Z"},"content_sha256":"3936d43beb2992eda2cf18bb3f4192be0662f81d0125c89776111b64ba167b0b","schema_version":"1.0","event_id":"sha256:3936d43beb2992eda2cf18bb3f4192be0662f81d0125c89776111b64ba167b0b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KLGNQ6LALRFRY6T2HEKVRQOZ6U/bundle.json","state_url":"https://pith.science/pith/KLGNQ6LALRFRY6T2HEKVRQOZ6U/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KLGNQ6LALRFRY6T2HEKVRQOZ6U/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-06T10:55:30Z","links":{"resolver":"https://pith.science/pith/KLGNQ6LALRFRY6T2HEKVRQOZ6U","bundle":"https://pith.science/pith/KLGNQ6LALRFRY6T2HEKVRQOZ6U/bundle.json","state":"https://pith.science/pith/KLGNQ6LALRFRY6T2HEKVRQOZ6U/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KLGNQ6LALRFRY6T2HEKVRQOZ6U/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:KLGNQ6LALRFRY6T2HEKVRQOZ6U","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"b9d9cb76a8ff0814afaca653609be609b2632f32c9b65e84c0f0bfa2adaad3a9","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-09T15:04:05Z","title_canon_sha256":"4414c5f9da24ca6125a4166338ad5dd24c02ba8be7fc88f8cd97af4a109ce738"},"schema_version":"1.0","source":{"id":"2607.08572","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.08572","created_at":"2026-07-10T01:19:53Z"},{"alias_kind":"arxiv_version","alias_value":"2607.08572v1","created_at":"2026-07-10T01:19:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.08572","created_at":"2026-07-10T01:19:53Z"},{"alias_kind":"pith_short_12","alias_value":"KLGNQ6LALRFR","created_at":"2026-07-10T01:19:53Z"},{"alias_kind":"pith_short_16","alias_value":"KLGNQ6LALRFRY6T2","created_at":"2026-07-10T01:19:53Z"},{"alias_kind":"pith_short_8","alias_value":"KLGNQ6LA","created_at":"2026-07-10T01:19:53Z"}],"graph_snapshots":[{"event_id":"sha256:3936d43beb2992eda2cf18bb3f4192be0662f81d0125c89776111b64ba167b0b","target":"graph","created_at":"2026-07-10T01:19:53Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2607.08572/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multimodal Large Language Models (MLLMs) often follow a fixed Think-then-Answer paradigm, which is inefficient in heterogeneous multitask settings because simple inputs may not require explicit reasoning while difficult ones can benefit substantially from it. Learning when to think is also unstable during post-training, where imbalanced rollouts can drive the model toward always-thinking or always-direct behavior. We propose Switch-Reasoner, a GRPO-based framework that learns to adaptively select reasoning modes for MLLMs. It treats thinking as a virtual tool invocation and allows the model to","authors_text":"Jian Liang, Jian Luan, Jinjie Li, Mang Ye, Pei Fu, Ruijie Luo, Shaojie Zhang, Wenke Huang, Yi R. Fung, Yiyang Fang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-09T15:04:05Z","title":"Switch-Reasoner: Learn When to Think in Multitask Mixtures via Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.08572","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:191e0df71db655b1d9affb9bbb8ad964a148b4898d7bae5b9549cb4c36399264","target":"record","created_at":"2026-07-10T01:19:53Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"b9d9cb76a8ff0814afaca653609be609b2632f32c9b65e84c0f0bfa2adaad3a9","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-09T15:04:05Z","title_canon_sha256":"4414c5f9da24ca6125a4166338ad5dd24c02ba8be7fc88f8cd97af4a109ce738"},"schema_version":"1.0","source":{"id":"2607.08572","kind":"arxiv","version":1}},"canonical_sha256":"52ccd879605c4b1c7a7a391558c1d9f51eb7e641b3b5bbdfa266c7bc905d50cb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"52ccd879605c4b1c7a7a391558c1d9f51eb7e641b3b5bbdfa266c7bc905d50cb","first_computed_at":"2026-07-10T01:19:53.932604Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-10T01:19:53.932604Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"F0lcHmFoSsBaKhOYOzUFWllLcf7XQbtiA1xzZ5HXSK+Xk33JsI/yBo7kTEnrlNghcwoQ1ODq3mbLozUx22NlAQ==","signature_status":"signed_v1","signed_at":"2026-07-10T01:19:53.932932Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.08572","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:191e0df71db655b1d9affb9bbb8ad964a148b4898d7bae5b9549cb4c36399264","sha256:3936d43beb2992eda2cf18bb3f4192be0662f81d0125c89776111b64ba167b0b"],"state_sha256":"77b29ff5301c863c55bd6b77cee5bb02fe798200fd56d7c8a8fde0eacc3cbae4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GIWvE7rIZ6BMQF82WOGQ7IK9/IMcsNTqWv7k4sTkN7zS+HVbvMOWM6g2lDmQFw2XcVMXGNXFOa140sweX1YGAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T10:55:30.539145Z","bundle_sha256":"d75c47c85e7903ec09498ab0a3165e0522ce2e5f745d0503b4467786acd26e97"}}