{"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/why-we-feel-breaking-boundaries-in-emotional","title":"Why We Feel: Breaking Boundaries in Emotional Reasoning with Multimodal Large Language Models","arxiv_id":"2504.07521","date":"2025-04-10","proceeding":null,"authors":["Yuxiang Lin","Jingdong Sun","Zhi-Qi Cheng","Jue Wang","Haomin Liang","Zebang Cheng","Yifei Dong","Jun-Yan He","Xiaojiang Peng","Xian-Sheng Hua"],"abstract":"Most existing emotion analysis emphasizes which emotion arises (e.g., happy, sad, angry) but neglects the deeper why. We propose Emotion Interpretation (EI), focusing on causal factors-whether explicit (e.g., observable objects, interpersonal interactions) or implicit (e.g., cultural context, off-screen events)-that drive emotional responses. Unlike traditional emotion recognition, EI tasks require reasoning about triggers instead of mere labeling. To facilitate EI research, we present EIBench, a large-scale benchmark encompassing 1,615 basic EI samples and 50 complex EI samples featuring multifaceted emotions. Each instance demands rationale-based explanations rather than straightforward categorization. We further propose a Coarse-to-Fine Self-Ask (CFSA) annotation pipeline, which guides Vision-Language Models (VLLMs) through iterative question-answer rounds to yield high-quality labels at scale. Extensive evaluations on open-source and proprietary large language models under four experimental settings reveal consistent performance gaps-especially for more intricate scenarios-underscoring EI's potential to enrich empathetic, context-aware AI applications. Our benchmark and methods are publicly available at: https://github.com/Lum1104/EIBench, offering a foundation for advanced multimodal causal analysis and next-generation affective computing.","url_abs":"https://arxiv.org/abs/2504.07521v2","url_pdf":"https://arxiv.org/pdf/2504.07521v2.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":"why-we-feel-breaking-boundaries-in-emotional","repo_url":"https://github.com/lum1104/eibench","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"why-we-feel-breaking-boundaries-in-emotional","repo_url":"https://github.com/Lum1104/MER-Factory","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"emotion-interpretation","task_name":"Emotion Interpretation"},{"task_slug":"emotion-recognition","task_name":"Emotion Recognition"}],"methods":[],"datasets_introduced":[{"slug":"eibench","name":"EIBench","full_name":""}],"methods_introduced":[],"results":[{"leaderboard":"/sota/emotion-interpretation-on-eibench","task":"Emotion Interpretation","dataset":"EIBench","model":"Claude-3-haiku","rank_in_archive_order":1,"of":13,"metrics":{"Recall":"63.24"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench","task":"Emotion Interpretation","dataset":"EIBench","model":"LLaVA-1.5 (13B)","rank_in_archive_order":2,"of":13,"metrics":{"Recall":"54.37"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench","task":"Emotion Interpretation","dataset":"EIBench","model":"LLaVA-NEXT (13B)","rank_in_archive_order":3,"of":13,"metrics":{"Recall":"54.33"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench","task":"Emotion Interpretation","dataset":"EIBench","model":"Claude-3-sonnet","rank_in_archive_order":4,"of":13,"metrics":{"Recall":"54.1"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench","task":"Emotion Interpretation","dataset":"EIBench","model":"LLaVA-NEXT (7B)","rank_in_archive_order":5,"of":13,"metrics":{"Recall":"53.82"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench","task":"Emotion Interpretation","dataset":"EIBench","model":"MiniGPT-v2","rank_in_archive_order":6,"of":13,"metrics":{"Recall":"52.89"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench","task":"Emotion Interpretation","dataset":"EIBench","model":"ChatGPT-4o","rank_in_archive_order":7,"of":13,"metrics":{"Recall":"49.99"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench","task":"Emotion Interpretation","dataset":"EIBench","model":"Video-LLaVA","rank_in_archive_order":8,"of":13,"metrics":{"Recall":"49.26"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench","task":"Emotion Interpretation","dataset":"EIBench","model":"LLaVA-NEXT (34B)","rank_in_archive_order":9,"of":13,"metrics":{"Recall":"49.03"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench","task":"Emotion Interpretation","dataset":"EIBench","model":"ChatGPT-4V","rank_in_archive_order":10,"of":13,"metrics":{"Recall":"46.86"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench","task":"Emotion Interpretation","dataset":"EIBench","model":"Otter","rank_in_archive_order":11,"of":13,"metrics":{"Recall":"42.81"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench","task":"Emotion Interpretation","dataset":"EIBench","model":"Qwen-vl-plus","rank_in_archive_order":12,"of":13,"metrics":{"Recall":"31"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench","task":"Emotion Interpretation","dataset":"EIBench","model":"Qwen-VL-Chat","rank_in_archive_order":13,"of":13,"metrics":{"Recall":"26.45"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench-complex","task":"Emotion Interpretation","dataset":"EIBench (complex)","model":"ChatGPT-4o","rank_in_archive_order":1,"of":13,"metrics":{"Recall":"39.27"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench-complex","task":"Emotion Interpretation","dataset":"EIBench (complex)","model":"LLaVA-NEXT (13B)","rank_in_archive_order":2,"of":13,"metrics":{"Recall":"39.16"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench-complex","task":"Emotion Interpretation","dataset":"EIBench (complex)","model":"LLaVA-NEXT (7B)","rank_in_archive_order":3,"of":13,"metrics":{"Recall":"38.71"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench-complex","task":"Emotion Interpretation","dataset":"EIBench (complex)","model":"LLaVA-1.5 (13B)","rank_in_archive_order":4,"of":13,"metrics":{"Recall":"38.1"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench-complex","task":"Emotion Interpretation","dataset":"EIBench (complex)","model":"LLaVA-NEXT (34B)","rank_in_archive_order":5,"of":13,"metrics":{"Recall":"35.37"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench-complex","task":"Emotion Interpretation","dataset":"EIBench (complex)","model":"MiniGPT-v2","rank_in_archive_order":6,"of":13,"metrics":{"Recall":"35.1"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench-complex","task":"Emotion Interpretation","dataset":"EIBench (complex)","model":"Video-LLaVA","rank_in_archive_order":7,"of":13,"metrics":{"Recall":"30.9"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench-complex","task":"Emotion Interpretation","dataset":"EIBench (complex)","model":"ChatGPT-4V","rank_in_archive_order":8,"of":13,"metrics":{"Recall":"28"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench-complex","task":"Emotion Interpretation","dataset":"EIBench (complex)","model":"Otter","rank_in_archive_order":9,"of":13,"metrics":{"Recall":"27.9"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench-complex","task":"Emotion Interpretation","dataset":"EIBench (complex)","model":"Claude-3-haiku","rank_in_archive_order":10,"of":13,"metrics":{"Recall":"24"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench-complex","task":"Emotion Interpretation","dataset":"EIBench (complex)","model":"Qwen-VL-Chat","rank_in_archive_order":11,"of":13,"metrics":{"Recall":"22"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench-complex","task":"Emotion Interpretation","dataset":"EIBench (complex)","model":"Claude-3-sonnet","rank_in_archive_order":12,"of":13,"metrics":{"Recall":"21.37"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-interpretation-on-eibench-complex","task":"Emotion Interpretation","dataset":"EIBench (complex)","model":"Qwen-vl-plus","rank_in_archive_order":13,"of":13,"metrics":{"Recall":"20.37"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2504.07521","atlas_url":"https://app.syntology.ai/?focus=2504.07521","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2504.07521"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/Lum1104/MER-Factory","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/lum1104/eibench","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":4,"ran_draft_wrong":1,"unverified":4},"by_repo_kind":{"official":{"samples":9,"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":"3842caa4b493f017","entry":"calculate_average_score","repo":"lum1104/eibench","repo_kind":"official","path":"EIBench/EI_Basic/get_scores.py","file_url":"https://github.com/lum1104/eibench/blob/HEAD/EIBench/EI_Basic/get_scores.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"3842caa4b493f017"}},{"code_sha256_prefix":"f41cb1a19b154297","entry":"encode_image","repo":"lum1104/eibench","repo_kind":"official","path":"EIBench/baselines/ChatGPT-4/gpt4-score-complex.py","file_url":"https://github.com/lum1104/eibench/blob/HEAD/EIBench/baselines/ChatGPT-4/gpt4-score-complex.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"f41cb1a19b154297"}},{"code_sha256_prefix":"03387f6b6c21f170","entry":"extract_scores_from_jsonl","repo":"lum1104/eibench","repo_kind":"official","path":"EIBench/EI_Basic/get_scores.py","file_url":"https://github.com/lum1104/eibench/blob/HEAD/EIBench/EI_Basic/get_scores.py","link_basis":"harvester_set","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":"03387f6b6c21f170"}},{"code_sha256_prefix":"4807f06cfd55795c","entry":"extract_scores_from_jsonl","repo":"lum1104/eibench","repo_kind":"official","path":"EIBench/EI_Complex/get_scores_complex.py","file_url":"https://github.com/lum1104/eibench/blob/HEAD/EIBench/EI_Complex/get_scores_complex.py","link_basis":"harvester_set","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":"4807f06cfd55795c"}},{"code_sha256_prefix":"c1a40518b7d33c75","entry":"get_ann","repo":"lum1104/eibench","repo_kind":"official","path":"EIBench/human_eval/web_ann_basic.py","file_url":"https://github.com/lum1104/eibench/blob/HEAD/EIBench/human_eval/web_ann_basic.py","link_basis":"harvester_set","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":"c1a40518b7d33c75"}},{"code_sha256_prefix":"d12384275cefed30","entry":"ask_chatgpt","repo":"lum1104/eibench","repo_kind":"official","path":"EIBench/EI_Basic/gpt-eval.py","file_url":"https://github.com/lum1104/eibench/blob/HEAD/EIBench/EI_Basic/gpt-eval.py","link_basis":"harvester_set","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":"d12384275cefed30"}},{"code_sha256_prefix":"6b60d87b77a0ce1c","entry":"ask_chatgpt","repo":"lum1104/eibench","repo_kind":"official","path":"EIBench/baselines/ChatGPT-4/gpt4-score-complex.py","file_url":"https://github.com/lum1104/eibench/blob/HEAD/EIBench/baselines/ChatGPT-4/gpt4-score-complex.py","link_basis":"harvester_set","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":"6b60d87b77a0ce1c"}},{"code_sha256_prefix":"e85a5515b1b5db8b","entry":"load_jsonl","repo":"lum1104/eibench","repo_kind":"official","path":"EIBench/human_eval/web_ann_basic.py","file_url":"https://github.com/lum1104/eibench/blob/HEAD/EIBench/human_eval/web_ann_basic.py","link_basis":"harvester_set","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":"e85a5515b1b5db8b"}},{"code_sha256_prefix":"e2378b5484a0e5e6","entry":"switch_label_based_on_dropdown","repo":"lum1104/eibench","repo_kind":"official","path":"EIBench/human_eval/web_ann_basic.py","file_url":"https://github.com/lum1104/eibench/blob/HEAD/EIBench/human_eval/web_ann_basic.py","link_basis":"harvester_set","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":"e2378b5484a0e5e6"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}