{"paper":{"title":"From Failure to Feedback: Group Revision Unlocks Hard Cases in Object-Level Grounding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Anjie Le, Can Peng, Fengbei Liu, Jiajun Deng, Jiayuan Zhu, Jiazhen Pan, Junde Wu, Yiping Ji, Yuyuan Liu","submitted_at":"2026-05-15T13:41:41Z","abstract_excerpt":"Finetuning Large Vision-Language Models with reinforcement learning has emerged as a promising approach to enhance their capability in object-level grounding. However, existing methods, mainly based on GRPO, assign rewards at the response level. Such sparse reward, often criterion-induced, leads to minimal learning signals when all candidate responses fail in challenging scenarios. In this work, we propose a group-revision optimisation paradigm that enhances learning on hard cases. It begins with a sampled initial response and generates a set of revised candidates to explore improved grounding"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.15951","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/2605.15951/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-19T17:33:44.880901Z","status":"skipped","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T17:01:55.713491Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"3b0782faac250be0b2ef5d856eba0687f6288359726966820024c091e9ad22a1"},"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"}