{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/noisyrollout-reinforcing-visual-reasoning","title":"NoisyRollout: Reinforcing Visual Reasoning with Data Augmentation","arxiv_id":"2504.13055","date":"2025-04-17","proceeding":null,"authors":["Xiangyan Liu","Jinjie Ni","Zijian Wu","Chao Du","Longxu Dou","Haonan Wang","Tianyu Pang","Michael Qizhe Shieh"],"abstract":"Recent advances in reinforcement learning (RL) have strengthened the reasoning capabilities of vision-language models (VLMs). However, enhancing policy exploration to more effectively scale test-time compute remains underexplored in VLMs. In addition, VLMs continue to struggle with imperfect visual perception, which in turn affects the subsequent reasoning process. To this end, we propose NoisyRollout, a simple yet effective RL approach that mixes trajectories from both clean and moderately distorted images to introduce targeted diversity in visual perception and the resulting reasoning patterns. Without additional training cost, NoisyRollout enhances the exploration capabilities of VLMs by incorporating a vision-oriented inductive bias. Furthermore, NoisyRollout employs a noise annealing schedule that gradually reduces distortion strength over training, ensuring benefit from noisy signals early while maintaining training stability and scalability in later stages. With just 2.1K training samples, NoisyRollout achieves state-of-the-art performance among open-source RL-tuned models on 5 out-of-domain benchmarks spanning both reasoning and perception tasks, while preserving comparable or even better in-domain performance.","url_abs":"https://arxiv.org/abs/2504.13055v1","url_pdf":"https://arxiv.org/pdf/2504.13055v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"noisyrollout-reinforcing-visual-reasoning","repo_url":"https://github.com/john-ai-lab/noisyrollout","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"data-augmentation","task_name":"Data Augmentation"},{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"inductive-bias","task_name":"Inductive Bias"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"visual-reasoning","task_name":"Visual Reasoning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2504.13055","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2504.13055"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/john-ai-lab/noisyrollout","reach":null}],"summary":{"ran":5,"unverified":1},"by_repo_kind":{"listed":{"samples":6,"ran":5,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"406bac274aa69cf3","entry":"FSDPConfig","repo":"john-ai-lab/noisyrollout","repo_kind":"listed","path":"verl/utils/image_aug.py","file_url":"https://github.com/john-ai-lab/noisyrollout/blob/HEAD/verl/utils/image_aug.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"406bac274aa69cf3"}},{"code_sha256_prefix":"b9dfdf0eb146ce26","entry":"ImageAugmenter","repo":"john-ai-lab/noisyrollout","repo_kind":"listed","path":"verl/utils/image_aug.py","file_url":"https://github.com/john-ai-lab/noisyrollout/blob/HEAD/verl/utils/image_aug.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b9dfdf0eb146ce26"}},{"code_sha256_prefix":"b84e254ea4a4e305","entry":"ModelConfig","repo":"john-ai-lab/noisyrollout","repo_kind":"listed","path":"verl/utils/image_aug.py","file_url":"https://github.com/john-ai-lab/noisyrollout/blob/HEAD/verl/utils/image_aug.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b84e254ea4a4e305"}},{"code_sha256_prefix":"d78dcf2efa0b656d","entry":"OffloadConfig","repo":"john-ai-lab/noisyrollout","repo_kind":"listed","path":"verl/utils/image_aug.py","file_url":"https://github.com/john-ai-lab/noisyrollout/blob/HEAD/verl/utils/image_aug.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"d78dcf2efa0b656d"}},{"code_sha256_prefix":"6567e140c9fbb8a1","entry":"OptimConfig","repo":"john-ai-lab/noisyrollout","repo_kind":"listed","path":"verl/utils/image_aug.py","file_url":"https://github.com/john-ai-lab/noisyrollout/blob/HEAD/verl/utils/image_aug.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"6567e140c9fbb8a1"}},{"code_sha256_prefix":"74ef2497323529e6","entry":"ActorConfig","repo":"john-ai-lab/noisyrollout","repo_kind":"listed","path":"verl/utils/image_aug.py","file_url":"https://github.com/john-ai-lab/noisyrollout/blob/HEAD/verl/utils/image_aug.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"74ef2497323529e6"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}