{"work":{"id":"a2dcce9c-5c33-49e2-989b-e0438645054f","openalex_id":null,"doi":null,"arxiv_id":"2509.26645","raw_key":null,"title":"TTT3R: 3D Reconstruction as Test-Time Training","authors":null,"authors_text":"Xingyu Chen, Yue Chen, Yuliang Xiu, Andreas Geiger, Anpei Chen","year":2025,"venue":"cs.CV","abstract":"Modern Recurrent Neural Networks have become a competitive architecture for 3D reconstruction due to their linear-time complexity. However, their performance degrades significantly when applied beyond the training context length, revealing limited length generalization. In this work, we revisit the 3D reconstruction foundation models from a Test-Time Training perspective, framing their designs as an online learning problem. Building on this perspective, we leverage the alignment confidence between the memory state and incoming observations to derive a closed-form learning rate for memory updates, to balance between retaining historical information and adapting to new observations. This training-free intervention, termed TTT3R, substantially improves length generalization, achieving a $2\\times$ improvement in global pose estimation over baselines, while operating at 20 FPS with just 6 GB of GPU memory to process thousands of images. Code is available in https://rover-xingyu.github.io/TTT3R","external_url":"https://arxiv.org/abs/2509.26645","cited_by_count":null,"metadata_source":"pith","metadata_fetched_at":"2026-05-25T04:35:20.798225+00:00","pith_arxiv_id":"2509.26645","created_at":"2026-05-09T06:25:48.637160+00:00","updated_at":"2026-05-25T04:35:20.798225+00:00","title_quality_ok":true,"display_title":"TTT3R: 3D Reconstruction as Test-Time Training","render_title":"TTT3R: 3D Reconstruction as Test-Time Training"},"hub":{"state":{"work_id":"a2dcce9c-5c33-49e2-989b-e0438645054f","tier":"hub","tier_reason":"10+ Pith inbound or 1,000+ external citations","pith_inbound_count":25,"external_cited_by_count":null,"distinct_field_count":2,"first_pith_cited_at":"2025-12-01T13:14:48+00:00","last_pith_cited_at":"2026-05-22T17:50:48+00:00","author_build_status":"not_needed","summary_status":"needed","contexts_status":"needed","graph_status":"needed","ask_index_status":"not_needed","reader_status":"not_needed","recognition_status":"not_needed","updated_at":"2026-05-27T21:37:58.632622+00:00","tier_text":"hub"},"tier":"hub","role_counts":[{"context_role":"background","n":6}],"polarity_counts":[{"context_polarity":"background","n":6}],"runs":{},"summary":{},"graph":{},"authors":[]}}