{"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/i-srt-aligning-large-multimodal-models-for","title":"ISR-DPO: Aligning Large Multimodal Models for Videos by Iterative Self-Retrospective DPO","arxiv_id":"2406.11280","date":"2024-06-17","proceeding":null,"authors":["Daechul Ahn","Yura Choi","San Kim","Youngjae Yu","Dongyeop Kang","Jonghyun Choi"],"abstract":"Iterative self-improvement, a concept extending beyond personal growth, has found powerful applications in machine learning, particularly in transforming weak models into strong ones. While recent advances in natural language processing have shown its efficacy through iterative preference optimization, applying this approach to Video Large Multi-modal Models (VLMMs) remains challenging due to modality misalignment. VLMMs struggle with this misalignment during iterative preference modeling, as the self-judge model often prioritizes linguistic knowledge over visual information. Additionally, iterative preference optimization can lead to visually hallucinated verbose responses due to length bias within the self-rewarding cycle. To address these issues, we propose Iterative Self-Retrospective Direct Preference Optimization (ISR-DPO), a method that uses self-retrospection to enhance preference modeling. This approach enhances the self-judge's focus on informative video regions, resulting in more visually grounded preferences. In extensive empirical evaluations across diverse video question answering benchmarks, the ISR-DPO significantly outperforms the state of the art. We are committed to open-sourcing our code, models, and datasets to encourage further investigation.","url_abs":"https://arxiv.org/abs/2406.11280v2","url_pdf":"https://arxiv.org/pdf/2406.11280v2.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":"i-srt-aligning-large-multimodal-models-for","repo_url":"https://github.com/snumprlab/SRT","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"i-srt-aligning-large-multimodal-models-for","repo_url":"https://github.com/snumprlab/isr-dpo","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"i-srt-aligning-large-multimodal-models-for","repo_url":"https://github.com/yonseivnl/vlm-rlaif","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"i-srt-aligning-large-multimodal-models-for","repo_url":"https://github.com/MindSpore-scientific-2/code-12/tree/main/isr","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"question-answering","task_name":"Question Answering"},{"task_slug":"response-generation","task_name":"Response Generation"},{"task_slug":"video-question-answering","task_name":"Video Question Answering"}],"methods":[{"method_slug":"focus","method_name":"Focus"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2406.11280","atlas_url":"https://app.syntology.ai/?focus=2406.11280","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2406.11280"}},"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/MindSpore-scientific-2/code-12/tree/main/isr","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/snumprlab/isr-dpo","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/yonseivnl/vlm-rlaif","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/snumprlab/SRT","reach":{"status":"ok"}}],"summary":{"ran_draft_wrong":8,"ran":11,"ran_violates":2,"unverified":9},"by_repo_kind":{"official":{"samples":15,"ran":9,"repositories":2},"listed":{"samples":15,"ran":12,"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":15,"samples":[{"code_sha256_prefix":"2154e44a6b59e432","entry":"batch_select","repo":"yonseivnl/vlm-rlaif","repo_kind":"listed","path":"RLAIF/models/reward_model.py","file_url":"https://github.com/yonseivnl/vlm-rlaif/blob/HEAD/RLAIF/models/reward_model.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":2,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"2154e44a6b59e432"}},{"code_sha256_prefix":"f1a2ea781d07a5ee","entry":"get_id_from_path","repo":"snumprlab/SRT","repo_kind":"official","path":"inference/inference.py","file_url":"https://github.com/snumprlab/SRT/blob/HEAD/inference/inference.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":2,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f1a2ea781d07a5ee"}},{"code_sha256_prefix":"a6989412aca721c1","entry":"unpack_dict","repo":"yonseivnl/vlm-rlaif","repo_kind":"listed","path":"RLAIF/models/reward_model.py","file_url":"https://github.com/yonseivnl/vlm-rlaif/blob/HEAD/RLAIF/models/reward_model.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":2,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"a6989412aca721c1"}},{"code_sha256_prefix":"5aa59c85970b28a7","entry":"all_gather_and_cat","repo":"yonseivnl/vlm-rlaif","repo_kind":"listed","path":"RLAIF/models/distributed_utils.py","file_url":"https://github.com/yonseivnl/vlm-rlaif/blob/HEAD/RLAIF/models/distributed_utils.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":"5aa59c85970b28a7"}},{"code_sha256_prefix":"d991ae45245789f9","entry":"alleq","repo":"yonseivnl/vlm-rlaif","repo_kind":"listed","path":"RLAIF/data_utils/common_utils.py","file_url":"https://github.com/yonseivnl/vlm-rlaif/blob/HEAD/RLAIF/data_utils/common_utils.py","link_basis":"first_harvest_node","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":"d991ae45245789f9"}},{"code_sha256_prefix":"af2735cffc6ee536","entry":"cast_with_native_amp","repo":"yonseivnl/vlm-rlaif","repo_kind":"listed","path":"RLAIF/models/rl_trainer.py","file_url":"https://github.com/yonseivnl/vlm-rlaif/blob/HEAD/RLAIF/models/rl_trainer.py","link_basis":"first_harvest_node","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":"af2735cffc6ee536"}},{"code_sha256_prefix":"c32b636f13ed9ee5","entry":"convert_to_scalar","repo":"snumprlab/SRT","repo_kind":"official","path":"trl/core.py","file_url":"https://github.com/snumprlab/SRT/blob/HEAD/trl/core.py","link_basis":"harvester_set","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":"c32b636f13ed9ee5"}},{"code_sha256_prefix":"42596ac916e883a5","entry":"create_scheduler","repo":"yonseivnl/vlm-rlaif","repo_kind":"listed","path":"RLAIF/models/trainer_utils.py","file_url":"https://github.com/yonseivnl/vlm-rlaif/blob/HEAD/RLAIF/models/trainer_utils.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":"42596ac916e883a5"}},{"code_sha256_prefix":"f5ff77db7757bc92","entry":"flatten_dict","repo":"yonseivnl/vlm-rlaif","repo_kind":"listed","path":"RLAIF/data_utils/common_utils.py","file_url":"https://github.com/yonseivnl/vlm-rlaif/blob/HEAD/RLAIF/data_utils/common_utils.py","link_basis":"first_harvest_node","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":"f5ff77db7757bc92"}},{"code_sha256_prefix":"42a46570620cd9fa","entry":"get_chunk","repo":"snumprlab/SRT","repo_kind":"official","path":"llava/eval/model_vqa.py","file_url":"https://github.com/snumprlab/SRT/blob/HEAD/llava/eval/model_vqa.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"well_formed","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"42a46570620cd9fa"}},{"code_sha256_prefix":"89ce33341af66037","entry":"get_last_checkpoint","repo":"yonseivnl/vlm-rlaif","repo_kind":"listed","path":"RLAIF/lora_utils.py","file_url":"https://github.com/yonseivnl/vlm-rlaif/blob/HEAD/RLAIF/lora_utils.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"89ce33341af66037"}},{"code_sha256_prefix":"bae18947b56f2be1","entry":"is_none","repo":"snumprlab/SRT","repo_kind":"official","path":"llava/eval/model_vqa_mmbench.py","file_url":"https://github.com/snumprlab/SRT/blob/HEAD/llava/eval/model_vqa_mmbench.py","link_basis":"harvester_set","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"bae18947b56f2be1"}},{"code_sha256_prefix":"63c0b0c7b99fe5b6","entry":"make_data","repo":"snumprlab/isr-dpo","repo_kind":"official","path":"Evaluation/eval/eval_qa.py","file_url":"https://github.com/snumprlab/isr-dpo/blob/HEAD/Evaluation/eval/eval_qa.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":"63c0b0c7b99fe5b6"}},{"code_sha256_prefix":"0f72ca36a1580010","entry":"maybe_truncate","repo":"snumprlab/isr-dpo","repo_kind":"official","path":"Evaluation/eval/eval_qa.py","file_url":"https://github.com/snumprlab/isr-dpo/blob/HEAD/Evaluation/eval/eval_qa.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"0f72ca36a1580010"}},{"code_sha256_prefix":"ddd3b7e8f7827aba","entry":"openai_generate","repo":"snumprlab/SRT","repo_kind":"official","path":"inference/chatgpt_inference.py","file_url":"https://github.com/snumprlab/SRT/blob/HEAD/inference/chatgpt_inference.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":"ddd3b7e8f7827aba"}},{"code_sha256_prefix":"606e2b3506d6ed7d","entry":"remove_pad_and_left_pad","repo":"yonseivnl/vlm-rlaif","repo_kind":"listed","path":"RLAIF/models/ppo_trainer.py","file_url":"https://github.com/yonseivnl/vlm-rlaif/blob/HEAD/RLAIF/models/ppo_trainer.py","link_basis":"first_harvest_node","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":"606e2b3506d6ed7d"}},{"code_sha256_prefix":"4354061aab23f5b9","entry":"remove_special_tokens","repo":"snumprlab/SRT","repo_kind":"official","path":"inference/inference.py","file_url":"https://github.com/snumprlab/SRT/blob/HEAD/inference/inference.py","link_basis":"harvester_set","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"4354061aab23f5b9"}},{"code_sha256_prefix":"4098e9c2afde4a5d","entry":"setup","repo":"yonseivnl/vlm-rlaif","repo_kind":"listed","path":"RLAIF/models/distributed_utils.py","file_url":"https://github.com/yonseivnl/vlm-rlaif/blob/HEAD/RLAIF/models/distributed_utils.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":"4098e9c2afde4a5d"}},{"code_sha256_prefix":"076c252c52cbb161","entry":"split_list","repo":"snumprlab/SRT","repo_kind":"official","path":"llava/eval/model_vqa.py","file_url":"https://github.com/snumprlab/SRT/blob/HEAD/llava/eval/model_vqa.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"076c252c52cbb161"}},{"code_sha256_prefix":"e5295408ef50f6e8","entry":"unpack_dict","repo":"yonseivnl/vlm-rlaif","repo_kind":"listed","path":"RLAIF/data_utils/common_utils.py","file_url":"https://github.com/yonseivnl/vlm-rlaif/blob/HEAD/RLAIF/data_utils/common_utils.py","link_basis":"first_harvest_node","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":"e5295408ef50f6e8"}},{"code_sha256_prefix":"9a0b92f09a2d6c76","entry":"whiten","repo":"yonseivnl/vlm-rlaif","repo_kind":"listed","path":"RLAIF/models/ppo_trainer.py","file_url":"https://github.com/yonseivnl/vlm-rlaif/blob/HEAD/RLAIF/models/ppo_trainer.py","link_basis":"first_harvest_node","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":"9a0b92f09a2d6c76"}},{"code_sha256_prefix":"a51ae45a1f6c0b5b","entry":"flatten_dict","repo":"snumprlab/SRT","repo_kind":"official","path":"trl/core.py","file_url":"https://github.com/snumprlab/SRT/blob/HEAD/trl/core.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":"a51ae45a1f6c0b5b"}},{"code_sha256_prefix":"f345a47ea5dbb5bc","entry":"get_id_from_frame_path","repo":"snumprlab/SRT","repo_kind":"official","path":"data_processing/utils.py","file_url":"https://github.com/snumprlab/SRT/blob/HEAD/data_processing/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":"f345a47ea5dbb5bc"}},{"code_sha256_prefix":"1f804baed33a101d","entry":"get_image","repo":"snumprlab/SRT","repo_kind":"official","path":"data_processing/utils.py","file_url":"https://github.com/snumprlab/SRT/blob/HEAD/data_processing/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":"1f804baed33a101d"}},{"code_sha256_prefix":"f0eed88e133ab3b9","entry":"get_peft_state_maybe_zero_3","repo":"yonseivnl/vlm-rlaif","repo_kind":"listed","path":"RLAIF/finetune_policy_init.py","file_url":"https://github.com/yonseivnl/vlm-rlaif/blob/HEAD/RLAIF/finetune_policy_init.py","link_basis":"first_harvest_node","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":"f0eed88e133ab3b9"}},{"code_sha256_prefix":"1c53657305b66e9f","entry":"get_peft_state_non_lora_maybe_zero_3","repo":"yonseivnl/vlm-rlaif","repo_kind":"listed","path":"RLAIF/finetune_policy_init.py","file_url":"https://github.com/yonseivnl/vlm-rlaif/blob/HEAD/RLAIF/finetune_policy_init.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":"1c53657305b66e9f"}},{"code_sha256_prefix":"5321a2aa0d7681cb","entry":"get_score","repo":"snumprlab/SRT","repo_kind":"official","path":"inference/utils.py","file_url":"https://github.com/snumprlab/SRT/blob/HEAD/inference/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":"5321a2aa0d7681cb"}},{"code_sha256_prefix":"6fac133e5d5b321b","entry":"load_frames","repo":"snumprlab/SRT","repo_kind":"official","path":"data_processing/utils.py","file_url":"https://github.com/snumprlab/SRT/blob/HEAD/data_processing/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":"6fac133e5d5b321b"}},{"code_sha256_prefix":"616ffbdc154ed2d8","entry":"maybe_zero_3","repo":"yonseivnl/vlm-rlaif","repo_kind":"listed","path":"RLAIF/finetune_policy_init.py","file_url":"https://github.com/yonseivnl/vlm-rlaif/blob/HEAD/RLAIF/finetune_policy_init.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":"616ffbdc154ed2d8"}},{"code_sha256_prefix":"085e70e0bd6e492d","entry":"stack_dicts","repo":"snumprlab/SRT","repo_kind":"official","path":"trl/core.py","file_url":"https://github.com/snumprlab/SRT/blob/HEAD/trl/core.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":"085e70e0bd6e492d"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}