{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:T7PRIAHHEGNOECVUL26FFSFZ4Z","short_pith_number":"pith:T7PRIAHH","schema_version":"1.0","canonical_sha256":"9fdf1400e7219ae20ab45ebc52c8b9e67fe8a9e37215a3affeadd29e38c0b516","source":{"kind":"arxiv","id":"2605.13375","version":1},"attestation_state":"computed","paper":{"title":"GRIP-VLM: Group-Relative Importance Pruning for Efficient Vision-Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"GRIP-VLM uses reinforcement learning to optimize discrete visual token pruning in VLMs, avoiding suboptimal local minima from gradient relaxations.","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Hao Wen, Liang Mi, Lichen Pang, Mingzhe Huang, Shansong Yang, Ting Cao, Weijun Wang, Xin Ding, Yuanchun Li, Yunxin Liu","submitted_at":"2026-05-13T11:32:03Z","abstract_excerpt":"In Vision-Language Models (VLMs), processing a massive number of visual tokens incurs prohibitive computational overhead. While recent training-aware pruning methods attempt to selectively discard redundant tokens, they largely rely on continuous-gradient relaxations. However, visual token pruning is inherently a discrete, non-convex combinatorial problem; consequently, these continuous approximations frequently trap the optimization in sub-optimal local minima, especially under aggressive compression budgets. To overcome this fundamental bottleneck, we propose GRIP-VLM, a Group-Relative Impor"},"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":true,"formal_links_present":false},"canonical_record":{"source":{"id":"2605.13375","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-13T11:32:03Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2d1a3380f6da5b8e97e08fba4ecd517b81b89b2a512b1ccbac00732ea21f7347","abstract_canon_sha256":"668b20bdedb2dd38cdba55b0c79ff2b87480be5eb068d63703fd1a8a3cecbe48"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T02:44:47.918241Z","signature_b64":"Cf2O5CKbpghp4mIRVY8tLzD73SuGIx1/1qj9KSy9mcJabh4Czkz6Wm/7NLI8P/vmtCD64AKRcNZAsqdwscXUDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9fdf1400e7219ae20ab45ebc52c8b9e67fe8a9e37215a3affeadd29e38c0b516","last_reissued_at":"2026-05-18T02:44:47.917854Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T02:44:47.917854Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GRIP-VLM: Group-Relative Importance Pruning for Efficient Vision-Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"GRIP-VLM uses reinforcement learning to optimize discrete visual token pruning in VLMs, avoiding suboptimal local minima from gradient relaxations.","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Hao Wen, Liang Mi, Lichen Pang, Mingzhe Huang, Shansong Yang, Ting Cao, Weijun Wang, Xin Ding, Yuanchun Li, Yunxin Liu","submitted_at":"2026-05-13T11:32:03Z","abstract_excerpt":"In Vision-Language Models (VLMs), processing a massive number of visual tokens incurs prohibitive computational overhead. While recent training-aware pruning methods attempt to selectively discard redundant tokens, they largely rely on continuous-gradient relaxations. However, visual token pruning is inherently a discrete, non-convex combinatorial problem; consequently, these continuous approximations frequently trap the optimization in sub-optimal local minima, especially under aggressive compression budgets. To overcome this fundamental bottleneck, we propose GRIP-VLM, a Group-Relative Impor"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"GRIP-VLM consistently outperforms heuristic and supervised-learning baselines, achieving a superior Pareto frontier and delivering up to a 15% inference speedup at equal accuracy.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the Group Relative Policy Optimization agent can reliably discover high-quality discrete pruning masks across varying compression budgets without retraining and without the instability typical of RL on combinatorial spaces.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"GRIP-VLM applies group-relative policy optimization via reinforcement learning to prune visual tokens in VLMs, yielding up to 15% inference speedup at matched accuracy over prior methods.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"GRIP-VLM uses reinforcement learning to optimize discrete visual token pruning in VLMs, avoiding suboptimal local minima from gradient relaxations.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"0f8705bb0e12e53541017e23ee834925ef3787f53d5c3f1cdc9cafa0005510e1"},"source":{"id":"2605.13375","kind":"arxiv","version":1},"verdict":{"id":"7199ddd1-4a63-4022-bb64-fbece73cbafa","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T19:12:10.099597Z","strongest_claim":"GRIP-VLM consistently outperforms heuristic and supervised-learning baselines, achieving a superior Pareto frontier and delivering up to a 15% inference speedup at equal accuracy.","one_line_summary":"GRIP-VLM applies group-relative policy optimization via reinforcement learning to prune visual tokens in VLMs, yielding up to 15% inference speedup at matched accuracy over prior methods.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the Group Relative Policy Optimization agent can reliably discover high-quality discrete pruning masks across varying compression budgets without retraining and without the instability typical of RL on combinatorial spaces.","pith_extraction_headline":"GRIP-VLM uses reinforcement learning to optimize discrete visual token pruning in VLMs, avoiding suboptimal local minima from gradient relaxations."},"references":{"count":42,"sample":[{"doi":"","year":2023,"title":"Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models","work_id":"4819f738-f69f-49dd-8bed-404f647de63a","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models","work_id":"a7e3a737-e007-42bc-be89-c4d34c5ee071","ref_index":2,"cited_arxiv_id":"2304.10592","is_internal_anchor":true},{"doi":"","year":2024,"title":"Improved baselines with visual instruction tuning","work_id":"1191d1e0-5e80-4225-a064-e1c04d70e44d","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Visual instruction tuning.Advances in neural information processing systems, 2024","work_id":"93eddd32-4e1d-4fb1-aa72-53eb41f6d1f2","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","work_id":"68a5177f-d644-44c1-bd4f-4e5278c22f5d","ref_index":5,"cited_arxiv_id":"2307.09288","is_internal_anchor":true}],"resolved_work":42,"snapshot_sha256":"a3d9548cde7690ae65e1aab4675a2c375d1d4f1d743b6dabd1d24314116a5b43","internal_anchors":12},"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":"2605.13375","created_at":"2026-05-18T02:44:47.917918+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.13375v1","created_at":"2026-05-18T02:44:47.917918+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13375","created_at":"2026-05-18T02:44:47.917918+00:00"},{"alias_kind":"pith_short_12","alias_value":"T7PRIAHHEGNO","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"T7PRIAHHEGNOECVU","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"T7PRIAHH","created_at":"2026-05-18T12:33:37.589309+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/T7PRIAHHEGNOECVUL26FFSFZ4Z","json":"https://pith.science/pith/T7PRIAHHEGNOECVUL26FFSFZ4Z.json","graph_json":"https://pith.science/api/pith-number/T7PRIAHHEGNOECVUL26FFSFZ4Z/graph.json","events_json":"https://pith.science/api/pith-number/T7PRIAHHEGNOECVUL26FFSFZ4Z/events.json","paper":"https://pith.science/paper/T7PRIAHH"},"agent_actions":{"view_html":"https://pith.science/pith/T7PRIAHHEGNOECVUL26FFSFZ4Z","download_json":"https://pith.science/pith/T7PRIAHHEGNOECVUL26FFSFZ4Z.json","view_paper":"https://pith.science/paper/T7PRIAHH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.13375&json=true","fetch_graph":"https://pith.science/api/pith-number/T7PRIAHHEGNOECVUL26FFSFZ4Z/graph.json","fetch_events":"https://pith.science/api/pith-number/T7PRIAHHEGNOECVUL26FFSFZ4Z/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/T7PRIAHHEGNOECVUL26FFSFZ4Z/action/timestamp_anchor","attest_storage":"https://pith.science/pith/T7PRIAHHEGNOECVUL26FFSFZ4Z/action/storage_attestation","attest_author":"https://pith.science/pith/T7PRIAHHEGNOECVUL26FFSFZ4Z/action/author_attestation","sign_citation":"https://pith.science/pith/T7PRIAHHEGNOECVUL26FFSFZ4Z/action/citation_signature","submit_replication":"https://pith.science/pith/T7PRIAHHEGNOECVUL26FFSFZ4Z/action/replication_record"}},"created_at":"2026-05-18T02:44:47.917918+00:00","updated_at":"2026-05-18T02:44:47.917918+00:00"}