{"url":"/task/multimodal-reasoning","name":"Multimodal Reasoning","slug":"multimodal-reasoning","description_markdown":"Reasoning over multimodal inputs.","categories":[{"name":"Computer Vision","url":"/area/computer-vision"},{"name":"Reasoning","url":"/area/reasoning"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":302,"papers_with_code":138,"benchmarks":3,"benchmark_tables_in_archive":3,"benchmark_tables_shown":3,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":11,"subtasks":1,"parent_tasks":0},"benchmarks":[{"leaderboard":"/sota/multimodal-reasoning-on-rebus","slug":"multimodal-reasoning-on-rebus","dataset":"REBUS","dataset_url":"/dataset/rebus","rows_in_archive":8,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"GPT-4V","paper_title":"REBUS: A Robust Evaluation Benchmark of Understanding Symbols","paper_url":"/paper/rebus-a-robust-evaluation-benchmark-of-1","paper_date":"2024-01-11","arxiv_id":"2401.05604","code_links":[{"title":"cvndsh/rebus","url":"https://github.com/cvndsh/rebus"}],"syntology":null}},{"leaderboard":"/sota/multimodal-reasoning-on-math-v","slug":"multimodal-reasoning-on-math-v","dataset":"MATH-V","dataset_url":"/dataset/math-v","rows_in_archive":4,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"GPT4V","paper_title":"Measuring Multimodal Mathematical Reasoning with MATH-Vision Dataset","paper_url":"/paper/measuring-multimodal-mathematical-reasoning","paper_date":"2024-02-22","arxiv_id":"2402.14804","code_links":[{"title":"mathllm/math-v","url":"https://github.com/mathllm/math-v"}],"syntology":null}},{"leaderboard":"/sota/multimodal-reasoning-on-algopuzzlevqa","slug":"multimodal-reasoning-on-algopuzzlevqa","dataset":"AlgoPuzzleVQA","dataset_url":"/dataset/algopuzzlevqa","rows_in_archive":1,"metrics":["Acc"],"first_row_in_archive_order":{"model":"GPT-4","paper_title":"Are Language Models Puzzle Prodigies? Algorithmic Puzzles Unveil Serious Challenges in Multimodal Reasoning","paper_url":"/paper/are-language-models-puzzle-prodigies","paper_date":"2024-03-06","arxiv_id":"2403.03864","code_links":[{"title":"declare-lab/puzzle-reasoning","url":"https://github.com/declare-lab/puzzle-reasoning"},{"title":"declare-lab/llm-puzzletest","url":"https://github.com/declare-lab/llm-puzzletest"}],"syntology":null}}],"datasets":[{"url":"/dataset/meetingbank","name":"MeetingBank","full_name":"","num_papers_in_archive":16},{"url":"/dataset/math-v","name":"MATH-V","full_name":"","num_papers_in_archive":12},{"url":"/dataset/vl-rewardbench","name":"VL-RewardBench","full_name":"","num_papers_in_archive":8},{"url":"/dataset/m3gia","name":"M3GIA","full_name":"","num_papers_in_archive":5},{"url":"/dataset/algopuzzlevqa","name":"AlgoPuzzleVQA","full_name":"","num_papers_in_archive":1},{"url":"/dataset/emma","name":"EMMA","full_name":"An Enhanced MultiModal ReAsoning Benchmark","num_papers_in_archive":1},{"url":"/dataset/gameqa","name":"GameQA","full_name":"GameQA-140K","num_papers_in_archive":1},{"url":"/dataset/polymath","name":"PolyMATH","full_name":"","num_papers_in_archive":1},{"url":"/dataset/r1-onevision","name":"R1-Onevision","full_name":"","num_papers_in_archive":1},{"url":"/dataset/rebus","name":"REBUS","full_name":"A Robust Evaluation Benchmark of Understanding Symbols","num_papers_in_archive":1},{"url":"/dataset/taste-music-dataset","name":"taste-music-dataset","full_name":"Taste Music Dataset","num_papers_in_archive":1}],"subtasks":[{"url":"/task/mme","name":"MME"}],"parent_tasks":[],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":30,"of":138,"tagged_in_all":302,"items":[{"url":"/paper/chartgalaxy-a-dataset-for-infographic-chart","title":"ChartGalaxy: A Dataset for Infographic Chart Understanding and Generation","date":"2025-05-24","arxiv_id":"2505.18668","repositories_listed":3,"syntology":{"n":3,"n_ran":3,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/webqa-multihop-and-multimodal-qa","title":"WebQA: Multihop and Multimodal QA","date":"2021-09-01","arxiv_id":"2109.00590","repositories_listed":3,"syntology":{"n":13,"n_ran":4,"n_unverified":9,"n_pointer_only":12}},{"url":"/paper/e-snli-ve-2-0-corrected-visual-textual","title":"e-SNLI-VE: Corrected Visual-Textual Entailment with Natural Language Explanations","date":"2020-04-07","arxiv_id":"2004.03744","repositories_listed":3,"syntology":{"n":3,"n_ran":3,"n_unverified":0,"n_pointer_only":2}},{"url":"/paper/skywork-r1v3-technical-report","title":"Skywork-R1V3 Technical Report","date":"2025-07-08","arxiv_id":"2507.06167","repositories_listed":2,"syntology":{"n":12,"n_ran":3,"n_unverified":9,"n_pointer_only":12}},{"url":"/paper/advancing-multimodal-reasoning-via","title":"Advancing Multimodal Reasoning via Reinforcement Learning with Cold Start","date":"2025-05-28","arxiv_id":"2505.22334","repositories_listed":2,"syntology":null},{"url":"/paper/emerging-properties-in-unified-multimodal","title":"Emerging Properties in Unified Multimodal Pretraining","date":"2025-05-20","arxiv_id":"2505.14683","repositories_listed":2,"syntology":{"n":21,"n_ran":9,"n_unverified":12,"n_pointer_only":0}},{"url":"/paper/mm-prm-enhancing-multimodal-mathematical","title":"MM-PRM: Enhancing Multimodal Mathematical Reasoning with Scalable Step-Level Supervision","date":"2025-05-19","arxiv_id":"2505.13427","repositories_listed":2,"syntology":{"n":2,"n_ran":2,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/sakura-on-the-multi-hop-reasoning-of-large","title":"SAKURA: On the Multi-hop Reasoning of Large Audio-Language Models Based on Speech and Audio Information","date":"2025-05-19","arxiv_id":"2505.13237","repositories_listed":2,"syntology":{"n":3,"n_ran":3,"n_unverified":0,"n_pointer_only":3}},{"url":"/paper/vl-rethinker-incentivizing-self-reflection-of","title":"VL-Rethinker: Incentivizing Self-Reflection of Vision-Language Models with Reinforcement Learning","date":"2025-04-10","arxiv_id":"2504.08837","repositories_listed":2,"syntology":{"n":12,"n_ran":2,"n_unverified":10,"n_pointer_only":0}},{"url":"/paper/the-jumping-reasoning-curve-tracking-the","title":"The Jumping Reasoning Curve? Tracking the Evolution of Reasoning Performance in GPT-[n] and o-[n] Models on Multimodal Puzzles","date":"2025-02-03","arxiv_id":"2502.01081","repositories_listed":2,"syntology":null},{"url":"/paper/llava-o1-let-vision-language-models-reason","title":"LLaVA-CoT: Let Vision Language Models Reason Step-by-Step","date":"2024-11-15","arxiv_id":"2411.10440","repositories_listed":2,"syntology":{"n":6,"n_ran":0,"n_unverified":6,"n_pointer_only":0}},{"url":"/paper/distill-visual-chart-reasoning-ability-from","title":"Distill Visual Chart Reasoning Ability from LLMs to MLLMs","date":"2024-10-24","arxiv_id":"2410.18798","repositories_listed":2,"syntology":{"n":13,"n_ran":0,"n_unverified":13,"n_pointer_only":13}},{"url":"/paper/2408-01337","title":"MuChoMusic: Evaluating Music Understanding in Multimodal Audio-Language Models","date":"2024-08-02","arxiv_id":"2408.01337","repositories_listed":2,"syntology":{"n":1,"n_ran":1,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/puzzlevqa-diagnosing-multimodal-reasoning","title":"PuzzleVQA: Diagnosing Multimodal Reasoning Challenges of Language Models with Abstract Visual Patterns","date":"2024-03-20","arxiv_id":"2403.13315","repositories_listed":2,"syntology":{"n":14,"n_ran":13,"n_unverified":1,"n_pointer_only":0}},{"url":"/paper/are-language-models-puzzle-prodigies","title":"Are Language Models Puzzle Prodigies? Algorithmic Puzzles Unveil Serious Challenges in Multimodal Reasoning","date":"2024-03-06","arxiv_id":"2403.03864","repositories_listed":2,"syntology":null},{"url":"/paper/mm-bigbench-evaluating-multimodal-models-on","title":"MM-BigBench: Evaluating Multimodal Models on Multimodal Content Comprehension Tasks","date":"2023-10-13","arxiv_id":"2310.09036","repositories_listed":2,"syntology":null},{"url":"/paper/beyond-chain-of-thought-effective-graph-of","title":"Beyond Chain-of-Thought, Effective Graph-of-Thought Reasoning in Language Models","date":"2023-05-26","arxiv_id":"2305.16582","repositories_listed":2,"syntology":{"n":2,"n_ran":2,"n_unverified":0,"n_pointer_only":2}},{"url":"/paper/multimodal-analogical-reasoning-over","title":"Multimodal Analogical Reasoning over Knowledge Graphs","date":"2022-10-01","arxiv_id":"2210.00312","repositories_listed":2,"syntology":{"n":27,"n_ran":7,"n_unverified":20,"n_pointer_only":0}},{"url":"/paper/dual-attention-networks-for-multimodal","title":"Dual Attention Networks for Multimodal Reasoning and Matching","date":"2016-11-02","arxiv_id":"1611.00471","repositories_listed":2,"syntology":null},{"url":"/paper/dreamvla-a-vision-language-action-model-1","title":"DreamVLA: A Vision-Language-Action Model Dreamed with Comprehensive World Knowledge","date":"2025-07-06","arxiv_id":"2507.04447","repositories_listed":1,"syntology":{"n":3,"n_ran":2,"n_unverified":1,"n_pointer_only":3}},{"url":"/paper/glm-4-1v-thinking-towards-versatile","title":"GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning","date":"2025-07-01","arxiv_id":"2507.01006","repositories_listed":1,"syntology":{"n":11,"n_ran":0,"n_unverified":11,"n_pointer_only":0}},{"url":"/paper/thinking-with-images-for-multimodal-reasoning","title":"Thinking with Images for Multimodal Reasoning: Foundations, Methods, and Future Frontiers","date":"2025-06-30","arxiv_id":"2506.23918","repositories_listed":1,"syntology":null},{"url":"/paper/humanomniv2-from-understanding-to-omni-modal","title":"HumanOmniV2: From Understanding to Omni-Modal Reasoning with Context","date":"2025-06-26","arxiv_id":"2506.21277","repositories_listed":1,"syntology":null},{"url":"/paper/machine-mental-imagery-empower-multimodal","title":"Machine Mental Imagery: Empower Multimodal Reasoning with Latent Visual Tokens","date":"2025-06-20","arxiv_id":"2506.17218","repositories_listed":1,"syntology":{"n":17,"n_ran":0,"n_unverified":17,"n_pointer_only":17}},{"url":"/paper/metis-rise-rl-incentivizes-and-sft-enhances","title":"Metis-RISE: RL Incentivizes and SFT Enhances Multimodal Reasoning Model Learning","date":"2025-06-16","arxiv_id":"2506.13056","repositories_listed":1,"syntology":null},{"url":"/paper/reinforcing-spatial-reasoning-in-vision","title":"Reinforcing Spatial Reasoning in Vision-Language Models with Interwoven Thinking and Visual Drawing","date":"2025-06-11","arxiv_id":"2506.09965","repositories_listed":1,"syntology":null},{"url":"/paper/play-to-generalize-learning-to-reason-through","title":"Play to Generalize: Learning to Reason Through Game Play","date":"2025-06-09","arxiv_id":"2506.08011","repositories_listed":1,"syntology":{"n":12,"n_ran":0,"n_unverified":12,"n_pointer_only":0}},{"url":"/paper/wethink-toward-general-purpose-vision","title":"WeThink: Toward General-purpose Vision-Language Reasoning via Reinforcement Learning","date":"2025-06-09","arxiv_id":"2506.07905","repositories_listed":1,"syntology":null},{"url":"/paper/learning-compact-vision-tokens-for-efficient","title":"Learning Compact Vision Tokens for Efficient Large Multimodal Models","date":"2025-06-08","arxiv_id":"2506.07138","repositories_listed":1,"syntology":null},{"url":"/paper/morse-500-a-programmatically-controllable","title":"MORSE-500: A Programmatically Controllable Video Benchmark to Stress-Test Multimodal Reasoning","date":"2025-06-05","arxiv_id":"2506.05523","repositories_listed":1,"syntology":null}],"syntology_records":18,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}