{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/improving-video-text-retrieval-by-multi","title":"Improving Video-Text Retrieval by Multi-Stream Corpus Alignment and Dual Softmax Loss","arxiv_id":"2109.04290","date":"2021-09-09","proceeding":null,"authors":["Xing Cheng","Hezheng Lin","Xiangyu Wu","Fan Yang","Dong Shen"],"abstract":"Employing large-scale pre-trained model CLIP to conduct video-text retrieval task (VTR) has become a new trend, which exceeds previous VTR methods. Though, due to the heterogeneity of structures and contents between video and text, previous CLIP-based models are prone to overfitting in the training phase, resulting in relatively poor retrieval performance. In this paper, we propose a multi-stream Corpus Alignment network with single gate Mixture-of-Experts (CAMoE) and a novel Dual Softmax Loss (DSL) to solve the two heterogeneity. The CAMoE employs Mixture-of-Experts (MoE) to extract multi-perspective video representations, including action, entity, scene, etc., then align them with the corresponding part of the text. In this stage, we conduct massive explorations towards the feature extraction module and feature alignment module. DSL is proposed to avoid the one-way optimum-match which occurs in previous contrastive methods. Introducing the intrinsic prior of each pair in a batch, DSL serves as a reviser to correct the similarity matrix and achieves the dual optimal match. DSL is easy to implement with only one-line code but improves significantly. The results show that the proposed CAMoE and DSL are of strong efficiency, and each of them is capable of achieving State-of-The-Art (SOTA) individually on various benchmarks such as MSR-VTT, MSVD, and LSMDC. Further, with both of them, the performance is advanced to a big extend, surpassing the previous SOTA methods for around 4.6\\% R@1 in MSR-VTT.","url_abs":"https://arxiv.org/abs/2109.04290v3","url_pdf":"https://arxiv.org/pdf/2109.04290v3.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":"improving-video-text-retrieval-by-multi","repo_url":"https://github.com/starmemda/CAMoE","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"improving-video-text-retrieval-by-multi","repo_url":"https://github.com/starmemda/camow","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"mixture-of-experts","task_name":"Mixture-of-Experts"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"text-retrieval","task_name":"Text Retrieval"},{"task_slug":"video-retrieval","task_name":"Video Retrieval"},{"task_slug":"video-text-retrieval","task_name":"Video-Text Retrieval"}],"methods":[{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"attention-dropout","method_name":"Attention Dropout"},{"method_slug":"bert","method_name":"BERT"},{"method_slug":"camoe","method_name":"CAMoE"},{"method_slug":"clip","method_name":"CLIP"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"dual-softmax-loss","method_name":"Dual Softmax Loss"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"linear-warmup-with-linear-decay","method_name":"Linear Warmup With Linear Decay"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"vision-transformer","method_name":"Vision Transformer"},{"method_slug":"weight-decay","method_name":"Weight Decay"},{"method_slug":"wordpiece","method_name":"WordPiece"}],"datasets_introduced":[],"methods_introduced":[{"slug":"camoe","name":"CAMoE","full_name":"CAMoE"}],"results":[{"leaderboard":"/sota/video-retrieval-on-activitynet","task":"Video Retrieval","dataset":"ActivityNet","model":"CAMoE","rank_in_archive_order":15,"of":31,"metrics":{"text-to-video Mean Rank":"6.3","text-to-video Median Rank":"1","text-to-video R@1":"51.0","text-to-video R@10":"87.6","text-to-video R@5":"77.7"},"uses_additional_data":true},{"leaderboard":"/sota/video-retrieval-on-didemo","task":"Video Retrieval","dataset":"DiDeMo","model":"CAMoE","rank_in_archive_order":32,"of":40,"metrics":{"text-to-video Mean Rank":"16.3","text-to-video Median Rank":"2.0","text-to-video R@1":"43.8","text-to-video R@10":"79.9","text-to-video R@5":"71.4","video-to-text Mean Rank":"10.2","video-to-text Median Rank":"2","video-to-text R@1":"45.5","video-to-text R@10":"80.5"},"uses_additional_data":true},{"leaderboard":"/sota/video-retrieval-on-lsmdc","task":"Video Retrieval","dataset":"LSMDC","model":"CAMoE","rank_in_archive_order":16,"of":38,"metrics":{"text-to-video Mean Rank":"54.4","text-to-video R@1":"25.9","text-to-video R@10":"53.7","text-to-video R@5":"46.1"},"uses_additional_data":true},{"leaderboard":"/sota/video-retrieval-on-msr-vtt","task":"Video Retrieval","dataset":"MSR-VTT","model":"CAMoE","rank_in_archive_order":22,"of":40,"metrics":{"text-to-video Mean Rank":"42.6","text-to-video Median Rank":"3","text-to-video R@1":"32.9","text-to-video R@10":"68.4","text-to-video R@5":"58.3","video-to-text Mean Rank":"3.8","video-to-text Median Rank":"1","video-to-text R@1":"59.8","video-to-text R@10":"92.8","video-to-text R@5":"86.2"},"uses_additional_data":true},{"leaderboard":"/sota/video-retrieval-on-msr-vtt-1ka","task":"Video Retrieval","dataset":"MSR-VTT-1kA","model":"CAMoE","rank_in_archive_order":23,"of":63,"metrics":{"text-to-video Mean Rank":"12.4","text-to-video Median Rank":"2","text-to-video R@1":"48.8","text-to-video R@10":"85.3","text-to-video R@5":"75.6","video-to-text Mean Rank":"9.9","video-to-text Median Rank":"2","video-to-text R@1":"50.3","video-to-text R@10":"83.8","video-to-text R@5":"74.6"},"uses_additional_data":true},{"leaderboard":"/sota/video-retrieval-on-msvd","task":"Video Retrieval","dataset":"MSVD","model":"CAMoE","rank_in_archive_order":9,"of":24,"metrics":{"text-to-video Mean Rank":"8.9","text-to-video Median Rank":"1","text-to-video R@1":"51.8","text-to-video R@10":"87.6","text-to-video R@5":"87.6","video-to-text Mean Rank":"3.1","video-to-text Median Rank":"1","video-to-text R@1":"69.3","video-to-text R@10":"94.6","video-to-text R@5":"90.6"},"uses_additional_data":true}],"syntology":{"syntology_url":"https://syntology.ai/paper/2109.04290","atlas_url":"https://app.syntology.ai/?focus=2109.04290","mcp":null,"developers":"https://syntology.ai/developers","held":true,"reason":"arxiv_id_in_atlas","checked_against":{"atlas_ids":549449,"atlas_ids_sha256":"afd72399dafd9906e9062927f70655952385a3d75488558e1f0484e950ab09b7","read_at":"2026-10-05T01:46:21Z"},"agent_calls":[{"tool":"get_citation_path","arguments":{"paper_1":"2109.04290"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"2109.04290\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"2109.04290"},"call":"get_concepts_for_paper(arxiv_id=\"2109.04290\")"}],"by_repo":[{"repo":"starmemda/CAMoE","kind":"official","samples":0,"ran":0,"instrument":0,"unverified":0},{"repo":"starmemda/camow","kind":"official","samples":0,"ran":0,"instrument":0,"unverified":0}],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"10ccd5552adf1d0b3632deb9689b5c793cbe6c01ba4f06a9dbad26ec7a2e2871","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}