{"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/arxiv-2606-08063","title":"Robust-U1: Can MLLMs Self-Recover Corrupted Visual Content for Robust Understanding?","arxiv_id":"2606.08063","date":"2026-06-06","proceeding":"ICML","authors":["Jiaqi Tang","Jianmin Chen","Youyang Zhai","Wei Wei","Runtao Liu","Mengjie Zhao","Xiangyu Wu","Qingfa Xiao","Qifeng Chen"],"abstract":"Multimodal Large Language Models (MLLMs) have demonstrated remarkable success in visual understanding, yet their performance degrades significantly under real-world visual corruptions. While existing robustness enhancement approaches exist, they are limited: black-box feature alignment lacks interpretability, and white-box text-based reasoning cannot restore lost pixel-level details. This work investigates a fundamental research question: Can MLLMs recover corrupted visual content by themselves? To address this, we propose Robust-U1, a novel framework that equips MLLMs with explicit visual self-recovery capability for robust understanding. The approach comprises three core stages: supervised fine-tuning for initial reconstruction, reinforcement learning with dual rewards (pixel-level SSIM and semantic-level CLIP similarity) for aligning high visual quality, and multimodal reasoning that jointly considers both the corrupted input and the recovered image. Extensive experiments demonstrate that Robust-U1 achieves state-of-the-art robustness on the real-world corruption benchmark and maintains superior performance under adversarial corruptions on general VQA benchmarks. Analysis confirms that high-quality visual recovery directly enhances reasoning performance, establishing self-recovery as a critical mechanism for robust visual understanding. The source code is available at https://github.com/jqtangust/Robust-U1.","url_abs":"https://arxiv.org/abs/2606.08063","url_pdf":"https://arxiv.org/pdf/2606.08063","source":{"archive":null,"snapshot":"2025-07-28","note":"not in the Papers with Code archive (frozen at the snapshot)","row_kind":"graph","title_abstract_authors_date":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)"},"code_links":[],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2606.08063","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2606.08063"}},"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":[{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/jqtangust/Robust-U1","reach":null}],"summary":{"ran_fixture":1,"ran":5,"ran_draft_wrong":1,"unverified":6},"by_repo_kind":{"found_in_text":{"samples":13,"ran":7,"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":13,"samples":[{"code_sha256_prefix":"e5947aba1d10885f","entry":"get_1d_sincos_pos_embed_from_grid","repo":"jqtangust/Robust-U1","repo_kind":"found_in_text","path":"modeling/modeling/bagel/modeling_utils.py","file_url":"https://github.com/jqtangust/Robust-U1/blob/HEAD/modeling/modeling/bagel/modeling_utils.py","link_basis":"plan_row","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"e5947aba1d10885f"}},{"code_sha256_prefix":"58f584dfd7b8fc3f","entry":"get_2d_sincos_pos_embed","repo":"jqtangust/Robust-U1","repo_kind":"found_in_text","path":"modeling/modeling/bagel/modeling_utils.py","file_url":"https://github.com/jqtangust/Robust-U1/blob/HEAD/modeling/modeling/bagel/modeling_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"58f584dfd7b8fc3f"}},{"code_sha256_prefix":"26da3ed73c66b411","entry":"pad_sequence","repo":"jqtangust/Robust-U1","repo_kind":"found_in_text","path":"modeling/modeling/bagel/qwen2_navit.py","file_url":"https://github.com/jqtangust/Robust-U1/blob/HEAD/modeling/modeling/bagel/qwen2_navit.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"26da3ed73c66b411"}},{"code_sha256_prefix":"6078af656ef97528","entry":"robust_tinyclip_score","repo":"jqtangust/Robust-U1","repo_kind":"found_in_text","path":"rewards/flow_grpo_adapter.py","file_url":"https://github.com/jqtangust/Robust-U1/blob/HEAD/rewards/flow_grpo_adapter.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"6078af656ef97528"}},{"code_sha256_prefix":"038b60d2c77bf8b2","entry":"run_image_edit","repo":"jqtangust/Robust-U1","repo_kind":"found_in_text","path":"inference/model_setup.py","file_url":"https://github.com/jqtangust/Robust-U1/blob/HEAD/inference/model_setup.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"038b60d2c77bf8b2"}},{"code_sha256_prefix":"dddcf5ce6a1048ab","entry":"save_reasoning_results","repo":"jqtangust/Robust-U1","repo_kind":"found_in_text","path":"inference/reasoner.py","file_url":"https://github.com/jqtangust/Robust-U1/blob/HEAD/inference/reasoner.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"dddcf5ce6a1048ab"}},{"code_sha256_prefix":"5fb036462ae82f9f","entry":"swish","repo":"jqtangust/Robust-U1","repo_kind":"found_in_text","path":"modeling/modeling/autoencoder.py","file_url":"https://github.com/jqtangust/Robust-U1/blob/HEAD/modeling/modeling/autoencoder.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"5fb036462ae82f9f"}},{"code_sha256_prefix":"665d8a4e8f673a4c","entry":"get_2d_sincos_pos_embed_from_grid","repo":"jqtangust/Robust-U1","repo_kind":"found_in_text","path":"modeling/modeling/bagel/modeling_utils.py","file_url":"https://github.com/jqtangust/Robust-U1/blob/HEAD/modeling/modeling/bagel/modeling_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"665d8a4e8f673a4c"}},{"code_sha256_prefix":"734993c851924b68","entry":"load_ae","repo":"jqtangust/Robust-U1","repo_kind":"found_in_text","path":"modeling/modeling/autoencoder.py","file_url":"https://github.com/jqtangust/Robust-U1/blob/HEAD/modeling/modeling/autoencoder.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"734993c851924b68"}},{"code_sha256_prefix":"996ec3da8f220bf2","entry":"restoration_score","repo":"jqtangust/Robust-U1","repo_kind":"found_in_text","path":"rewards/flow_grpo_adapter.py","file_url":"https://github.com/jqtangust/Robust-U1/blob/HEAD/rewards/flow_grpo_adapter.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"996ec3da8f220bf2"}},{"code_sha256_prefix":"fc21f3bf737b62db","entry":"robust_structure_score","repo":"jqtangust/Robust-U1","repo_kind":"found_in_text","path":"rewards/flow_grpo_adapter.py","file_url":"https://github.com/jqtangust/Robust-U1/blob/HEAD/rewards/flow_grpo_adapter.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"fc21f3bf737b62db"}},{"code_sha256_prefix":"0e9a007209e3e65c","entry":"structure_score","repo":"jqtangust/Robust-U1","repo_kind":"found_in_text","path":"rewards/rewards.py","file_url":"https://github.com/jqtangust/Robust-U1/blob/HEAD/rewards/rewards.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"0e9a007209e3e65c"}},{"code_sha256_prefix":"df4cde9b7ea87f02","entry":"tinyclip_score","repo":"jqtangust/Robust-U1","repo_kind":"found_in_text","path":"rewards/rewards.py","file_url":"https://github.com/jqtangust/Robust-U1/blob/HEAD/rewards/rewards.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"df4cde9b7ea87f02"}}]},"arxiv_metadata":{"licence":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)","fields":["title","abstract","authors","date"],"primary_category":"cs.CL","source":"arxiv_2026.jsonl"},"syntology_extracted_results":null}