{"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/large-language-models-are-temporal-and-causal","title":"Large Language Models are Temporal and Causal Reasoners for Video Question Answering","arxiv_id":"2310.15747","date":"2023-10-24","proceeding":null,"authors":["Dohwan Ko","Ji Soo Lee","Wooyoung Kang","Byungseok Roh","Hyunwoo J. Kim"],"abstract":"Large Language Models (LLMs) have shown remarkable performances on a wide range of natural language understanding and generation tasks. We observe that the LLMs provide effective priors in exploiting $\\textit{linguistic shortcuts}$ for temporal and causal reasoning in Video Question Answering (VideoQA). However, such priors often cause suboptimal results on VideoQA by leading the model to over-rely on questions, $\\textit{i.e.}$, $\\textit{linguistic bias}$, while ignoring visual content. This is also known as `ungrounded guesses' or `hallucinations'. To address this problem while leveraging LLMs' prior on VideoQA, we propose a novel framework, Flipped-VQA, encouraging the model to predict all the combinations of $\\langle$V, Q, A$\\rangle$ triplet by flipping the source pair and the target label to understand their complex relationships, $\\textit{i.e.}$, predict A, Q, and V given a VQ, VA, and QA pairs, respectively. In this paper, we develop LLaMA-VQA by applying Flipped-VQA to LLaMA, and it outperforms both LLMs-based and non-LLMs-based models on five challenging VideoQA benchmarks. Furthermore, our Flipped-VQA is a general framework that is applicable to various LLMs (OPT and GPT-J) and consistently improves their performances. We empirically demonstrate that Flipped-VQA not only enhances the exploitation of linguistic shortcuts but also mitigates the linguistic bias, which causes incorrect answers over-relying on the question. Code is available at https://github.com/mlvlab/Flipped-VQA.","url_abs":"https://arxiv.org/abs/2310.15747v2","url_pdf":"https://arxiv.org/pdf/2310.15747v2.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":"large-language-models-are-temporal-and-causal","repo_url":"https://github.com/mlvlab/Flipped-VQA","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"natural-language-understanding","task_name":"Natural Language Understanding"},{"task_slug":"question-answering","task_name":"Question Answering"},{"task_slug":null,"task_name":"Triplet"},{"task_slug":"video-question-answering","task_name":"Video Question Answering"},{"task_slug":"visual-question-answering","task_name":"Visual Question Answering (VQA)"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/video-question-answering-on-dramaqa","task":"Video Question Answering","dataset":"DramaQA","model":"LLaMA-VQA","rank_in_archive_order":1,"of":1,"metrics":{"Accuracy":"84.1"},"uses_additional_data":false},{"leaderboard":"/sota/video-question-answering-on-next-qa","task":"Video Question Answering","dataset":"NExT-QA","model":"LLaMA-VQA (33B)","rank_in_archive_order":24,"of":47,"metrics":{"Accuracy":"75.5"},"uses_additional_data":false},{"leaderboard":"/sota/video-question-answering-on-situated","task":"Video Question Answering","dataset":"STAR Benchmark","model":"LLaMA-VQA","rank_in_archive_order":2,"of":17,"metrics":{"Average Accuracy":"65.4"},"uses_additional_data":false},{"leaderboard":"/sota/video-question-answering-on-tvqa","task":"Video Question Answering","dataset":"TVQA","model":"LLaMA-VQA","rank_in_archive_order":1,"of":6,"metrics":{"Accuracy":"82.2"},"uses_additional_data":false},{"leaderboard":"/sota/video-question-answering-on-vlep","task":"Video Question Answering","dataset":"VLEP","model":"LLaMA-VQA","rank_in_archive_order":1,"of":1,"metrics":{"Accuracy":"71.0"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2310.15747","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2310.15747"}},"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":"deterministic:regex_extraction","url":"https://github.com/mlvlab/Flipped-VQA","reach":null}],"summary":{"unverified":4},"by_repo_kind":{"official":{"samples":4,"ran":0,"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":"ea4fa90ff313cb77","entry":"Attention","repo":"mlvlab/Flipped-VQA","repo_kind":"official","path":"llama/model.py","file_url":"https://github.com/mlvlab/Flipped-VQA/blob/HEAD/llama/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ea4fa90ff313cb77"}},{"code_sha256_prefix":"6f671e3b713c8c1b","entry":"ModelArgs","repo":"mlvlab/Flipped-VQA","repo_kind":"official","path":"llama/model.py","file_url":"https://github.com/mlvlab/Flipped-VQA/blob/HEAD/llama/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6f671e3b713c8c1b"}},{"code_sha256_prefix":"afd800d945fafd04","entry":"Transformer","repo":"mlvlab/Flipped-VQA","repo_kind":"official","path":"llama/model.py","file_url":"https://github.com/mlvlab/Flipped-VQA/blob/HEAD/llama/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"afd800d945fafd04"}},{"code_sha256_prefix":"ed5543f646686967","entry":"TransformerBlock","repo":"mlvlab/Flipped-VQA","repo_kind":"official","path":"llama/model.py","file_url":"https://github.com/mlvlab/Flipped-VQA/blob/HEAD/llama/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ed5543f646686967"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}