{"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/temple-temporal-preference-learning-of-video","title":"TEMPLE:Temporal Preference Learning of Video LLMs via Difficulty Scheduling and Pre-SFT Alignment","arxiv_id":"2503.16929","date":"2025-03-21","proceeding":null,"authors":["Shicheng Li","Lei LI","Kun Ouyang","Shuhuai Ren","Yuanxin Liu","Yuanxing Zhang","Fuzheng Zhang","Lingpeng Kong","Qi Liu","Xu sun"],"abstract":"Video Large Language Models (Video LLMs) have achieved significant success by leveraging a two-stage paradigm: pretraining on large-scale video-text data for vision-language alignment, followed by supervised fine-tuning (SFT) for task-specific capabilities. However, existing approaches struggle with temporal reasoning due to weak temporal correspondence in the data and reliance on the next-token prediction paradigm during training. To address these limitations, we propose TEMPLE (TEMporal Preference Learning), a systematic framework that enhances Video LLMs' temporal reasoning capabilities through Direct Preference Optimization (DPO). To facilitate this, we introduce an automated preference data generation pipeline that systematically constructs preference pairs by selecting videos that are rich in temporal information, designing video-specific perturbation strategies, and finally evaluating model responses on clean and perturbed video inputs. Our temporal alignment features two key innovations: curriculum learning which that progressively increases perturbation difficulty to improve model robustness and adaptability; and \"Pre-SFT Alignment'', applying preference optimization before instruction tuning to prioritize fine-grained temporal comprehension. Extensive experiments demonstrate that our approach consistently improves Video LLM performance across multiple benchmarks with a relatively small set of self-generated DPO data. We further analyze the transferability of DPO data across architectures and the role of difficulty scheduling in optimization. Our findings highlight our TEMPLE as a scalable and efficient complement to SFT-based methods, paving the way for developing reliable Video LLMs. Code is available at https://github.com/lscpku/TEMPLE.","url_abs":"https://arxiv.org/abs/2503.16929v2","url_pdf":"https://arxiv.org/pdf/2503.16929v2.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":"temple-temporal-preference-learning-of-video","repo_url":"https://github.com/lscpku/temple","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"scheduling","task_name":"Scheduling"}],"methods":[{"method_slug":"dpo","method_name":"DPO"},{"method_slug":"set","method_name":"SET"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2503.16929","atlas_url":"https://app.syntology.ai/?focus=2503.16929","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2503.16929"}},"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/lscpku/temple","reach":null}],"summary":{"ran_draft_wrong":1,"unverified":1},"by_repo_kind":{"official":{"samples":2,"ran":1,"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":2,"samples":[{"code_sha256_prefix":"db41886fef3918ad","entry":"load_jsonl","repo":"lscpku/temple","repo_kind":"official","path":"preprocess.py","file_url":"https://github.com/lscpku/temple/blob/HEAD/preprocess.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"db41886fef3918ad"}},{"code_sha256_prefix":"bd95f51056c675b9","entry":"process","repo":"lscpku/temple","repo_kind":"official","path":"preprocess.py","file_url":"https://github.com/lscpku/temple/blob/HEAD/preprocess.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":"bd95f51056c675b9"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}