{"paper":{"id":"2610.21114","date":"2026-10-22","title":"Spectral-Spatial Interpretation","author_names":["Haotian Ma","Ruqi Yang","Philip Townsend"],"github":"https://github.com/AtheavenMa/Spectral-Spatial-Interpretation","project_page":"https://github.com/AtheavenMa/Spectral-Spatial-Interpretation","upvotes":757,"comments":2,"peak":757,"first_day":"2026-10-22","up_d0":54,"up_d1":136,"up_d3":685,"up_d7":757,"purged":0,"up_close":757,"closed":true,"week":"2026-W43","month":"2026-10","early":false,"rank_day":2,"n_day":58,"rank_week":2,"n_week":312,"rank_month":2,"n_month":1363,"summary":"Post-hoc image explanations usually show where a model looks using spatial heatmaps or indicate which channels matter using global channel scores. These separate views are limiting for multispectral data because channel relevance is often spatially localized and physically meaningful. We introduce Spectral-Spatial Interpretation (SSI), a full-grid attribution method over position-channel players (g,c). For each spatial position, SSI fixes the observed surrounding context, defines a conditional game over the channels at that position, and applies the Shapley value to this local game. The result is a full spectral-spatial attribution matrix rather than a single heatmap or a global channel ranking. SSI has a clear local interpretation. It satisfies Shapley fairness within each position, and aggregating its channel attributions exactly recovers the position-level contextual contrast obtained by replacing that position with the baseline. SSI also differs from context-averaged Owen-style hierarchical accounting, which uses the same position-channel grouping but averages over external spatial coalitions. On HLS Burn Scar, SSI outperforms an MC-Owen estimator across retained attribution metrics while using only about 6% of its evaluation budget. Across CUB-200, EuroSAT, and HLS Burn Scar under both per-channel mean and zero baselines, SSI achieves the best joint deletion and insertion AUC over dense, flat, and rank-1 baselines, yields the strongest spectral hit rates and channel deletion, and leads spatial localization on both annotated datasets. Code is available at https://github.com/AtheavenMa/Spectral-Spatial-Interpretation.","authors":[{"name":"Haotian Ma","user":"Hunter26666666","fullname":"Haotian Ma","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6a1143c9b5cb600b5020d89b/GdmGt1ADypgvgMZnWiXXP.png","status":"claimed_verified","hidden":false},{"name":"Ruqi Yang","user":null,"fullname":null,"avatar":null,"status":null,"hidden":false},{"name":"Philip Townsend","user":null,"fullname":null,"avatar":null,"status":null,"hidden":false}],"org":null,"org_fullname":null,"org_avatar":null,"submitter":"Hunter26666666","submitter_fullname":"Haotian Ma","submitter_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6a1143c9b5cb600b5020d89b/GdmGt1ADypgvgMZnWiXXP.png","github_stars":51,"published_at":"2026-10-15T00:00:00.000Z","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.21114.png","author_participating":true,"used_models":0,"used_dl30":0},"series":{"day":["2026-10-22","2026-10-23","2026-10-24","2026-10-25","2026-10-26","2026-10-27","2026-10-28","2026-10-29"],"upvotes":[54,136,542,685,754,757,757,757],"comments":[1,2,2,2,2,2,2,2]},"rank":{"day":["2026-10-22","2026-10-23","2026-10-24","2026-10-25","2026-10-26","2026-10-27","2026-10-28","2026-10-29"],"rank":[8,6,1,1,1,1,2,2]},"purges":[],"neighbours":[{"id":"2610.28402","date":"2026-10-22","title":"StreamEval: Evaluating Interactive Streaming Video Generation Models","upvotes":864,"up_close":864,"up_d0":345,"up_d1":430,"comments":2,"peak":864,"purged":0,"closed":true,"early":false,"rank_day":1,"n_day":58,"github":"https://github.com/StreamEval/StreamEval","github_stars":523,"org":null,"org_fullname":null,"org_avatar":null,"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.28402.png","used_models":0,"used_dl30":0},{"id":"2610.21114","date":"2026-10-22","title":"Spectral-Spatial Interpretation","upvotes":757,"up_close":757,"up_d0":54,"up_d1":136,"comments":2,"peak":757,"purged":0,"closed":true,"early":false,"rank_day":2,"n_day":58,"github":"https://github.com/AtheavenMa/Spectral-Spatial-Interpretation","github_stars":51,"org":null,"org_fullname":null,"org_avatar":null,"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.21114.png","used_models":0,"used_dl30":0},{"id":"2610.11942","date":"2026-10-22","title":"Right Screen, Wrong Transition: World Models as Verifiers for GUI Agents","upvotes":189,"up_close":189,"up_d0":83,"up_d1":115,"comments":2,"peak":189,"purged":0,"closed":true,"early":false,"rank_day":3,"n_day":58,"github":"https://github.com/jiamingzhang94/lgwm","github_stars":4,"org":null,"org_fullname":null,"org_avatar":null,"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.11942.png","used_models":0,"used_dl30":0},{"id":"2610.29989","date":"2026-10-22","title":"Looking Ahead: Modeling Future Evolution in Autoregressive Video Generation","upvotes":174,"up_close":174,"up_d0":110,"up_d1":158,"comments":2,"peak":174,"purged":0,"closed":true,"early":false,"rank_day":4,"n_day":58,"github":"https://github.com/TencentARC/Looking-Ahead","github_stars":145,"org":"TencentARC","org_fullname":"ARC Lab, Tencent","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/1625552871844-60e272ca6c78a8c122b12127.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.29989.png","used_models":0,"used_dl30":0},{"id":"2610.26659","date":"2026-10-22","title":"SpatialScout: Can VLM Agents Actively Explore and Map 3D Scenes?","upvotes":170,"up_close":170,"up_d0":129,"up_d1":147,"comments":3,"peak":170,"purged":0,"closed":true,"early":false,"rank_day":5,"n_day":58,"github":"https://github.com/JesseZZZZZ/SpatialScout","github_stars":12,"org":"PekingUniversity","org_fullname":"Peking University","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/vavgrBsnkSejriUF4lXDE.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.26659.png","used_models":0,"used_dl30":0},{"id":"2610.29997","date":"2026-10-22","title":"DONIC: Generalist Dexterous Bimanual Tracking Control from Human Demonstrations","upvotes":162,"up_close":162,"up_d0":124,"up_d1":149,"comments":2,"peak":162,"purged":0,"closed":true,"early":false,"rank_day":6,"n_day":58,"github":"https://github.com/Ropedia/DONIC","github_stars":51,"org":"ropedia-ai","org_fullname":"Ropedia","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/69b116c65a198623dcbcc950/QFbPRKJTaUrMeF_rk7_4x.jpeg","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.29997.png","used_models":0,"used_dl30":0},{"id":"2610.28577","date":"2026-10-22","title":"Cache Engineering for Agents","upvotes":161,"up_close":161,"up_d0":112,"up_d1":139,"comments":1,"peak":161,"purged":0,"closed":true,"early":false,"rank_day":7,"n_day":58,"github":"https://github.com/thu-nics/AKV","github_stars":155,"org":"nics-efc","org_fullname":"Tsinghua-NICS-EFC","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/641031b1a78453b8d96b8420/vmgxct2WsyHcKT2x2NAYT.jpeg","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.28577.png","used_models":0,"used_dl30":0},{"id":"2610.26828","date":"2026-10-22","title":"Agent World Model in the Wild: Computer-Use Agent Environment Synthesis","upvotes":101,"up_close":101,"up_d0":64,"up_d1":77,"comments":2,"peak":101,"purged":0,"closed":true,"early":false,"rank_day":8,"n_day":58,"github":"https://github.com/MSR-Orchard/WildAWM","github_stars":82,"org":"MicrosoftResearch","org_fullname":"Microsoft Research","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6529a4f2f1205983224fa513/PeuVr7jSuJflmDBBGxoDX.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.26828.png","used_models":0,"used_dl30":0},{"id":"2610.28698","date":"2026-10-22","title":"Scaling in Games: Continued Pre-Training For Embodied Agents In Diverse Virtual Worlds","upvotes":86,"up_close":86,"up_d0":47,"up_d1":60,"comments":2,"peak":86,"purged":0,"closed":true,"early":false,"rank_day":9,"n_day":58,"github":"https://github.com/THUSI-Lab/GameScaling","github_stars":431,"org":"THUSI-Lab","org_fullname":"Tsinghua Spatial Intelligence Lab","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/669a6a4a7f28c43e09b7a6f5/0K36706OPoFd7oKA9p00w.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.28698.png","used_models":0,"used_dl30":0},{"id":"2610.20538","date":"2026-10-22","title":"PrismaCoder: Pushing the Boundaries of Open Multimodal Code Intelligence through Decomposed Rubric Synthesis","upvotes":72,"up_close":72,"up_d0":49,"up_d1":60,"comments":1,"peak":72,"purged":0,"closed":true,"early":false,"rank_day":10,"n_day":58,"github":"https://github.com/QiushiSun/PrismaCoder","github_stars":44,"org":"hkunlp","org_fullname":"NLP Group of The University of Hong Kong","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/1639647572687-618767e4238063b4615d042b.jpeg","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.20538.png","used_models":0,"used_dl30":0},{"id":"2610.29722","date":"2026-10-22","title":"Retrieval Augmented Generation Compositionality","upvotes":55,"up_close":55,"up_d0":27,"up_d1":40,"comments":1,"peak":55,"purged":0,"closed":true,"early":false,"rank_day":11,"n_day":58,"github":"https://github.com/dolev31/lasagna","github_stars":23,"org":"ibm","org_fullname":"IBM","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/637bfdf60dc13843b468ac20/9228luWRoGbZwKGxkOOsj.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.29722.png","used_models":0,"used_dl30":0},{"id":"2610.29680","date":"2026-10-22","title":"TraceForge: Distilling with Your Own Thinking","upvotes":50,"up_close":50,"up_d0":19,"up_d1":34,"comments":2,"peak":50,"purged":0,"closed":true,"early":false,"rank_day":12,"n_day":58,"github":null,"github_stars":null,"org":"IndependentResearcher121","org_fullname":"Independent Research","org_avatar":"https://www.gravatar.com/avatar/bbcac9111139876b6ff73360af68667d?d=retro&size=100","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.29680.png","used_models":0,"used_dl30":0},{"id":"2610.17860","date":"2026-10-22","title":"Socratic Skill Learning: Interleaving Teaching and Practice for Transferable Agent Skills","upvotes":49,"up_close":49,"up_d0":30,"up_d1":40,"comments":2,"peak":49,"purged":0,"closed":true,"early":false,"rank_day":13,"n_day":58,"github":"https://github.com/NJU-LINK/Socratic-Skill-Learning","github_stars":2,"org":"NJU-LINK","org_fullname":"NJU-LINK Lab","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/67f9d060395fb1a0d7e4ae21/O3V4UZjcSGnOivcQqTcXW.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.17860.png","used_models":0,"used_dl30":0},{"id":"2610.28589","date":"2026-10-22","title":"AgenticRoboRL: Bringing Coding-Agent Post-Training to Agentic Robotics","upvotes":49,"up_close":49,"up_d0":21,"up_d1":37,"comments":2,"peak":49,"purged":0,"closed":true,"early":false,"rank_day":13,"n_day":58,"github":"https://github.com/xinxinliu-ai/AgenticRoboRL","github_stars":48,"org":"UCF-Ai-LLMs","org_fullname":"University of Central Florida","org_avatar":"https://www.gravatar.com/avatar/6fb0661af6e4ec5ded3585562f94bbce?d=retro&size=100","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.28589.png","used_models":0,"used_dl30":0},{"id":"2610.28887","date":"2026-10-22","title":"Building AI Research Teams for Autonomous Discovery of Linear Attention Architectures","upvotes":45,"up_close":45,"up_d0":16,"up_d1":25,"comments":2,"peak":45,"purged":0,"closed":true,"early":false,"rank_day":15,"n_day":58,"github":null,"github_stars":null,"org":"UCSantaBarbara","org_fullname":"University of California, Santa Barbara","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/68e396f2b5bb631e9b2fac9a/AoPyKoRF9O8PHOoaJOi7h.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.28887.png","used_models":0,"used_dl30":0},{"id":"2610.29968","date":"2026-10-22","title":"CulturalRiddles: A Multicultural Benchmark of Riddles","upvotes":45,"up_close":45,"up_d0":15,"up_d1":32,"comments":2,"peak":45,"purged":0,"closed":true,"early":false,"rank_day":15,"n_day":58,"github":"https://github.com/Cohere-Labs-Community/CulturalRiddles","github_stars":12,"org":"CohereLabsCommunity","org_fullname":"Cohere Labs Community","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6351689040dffad651f0d26b/_t7IaE_wZy2SLYmQo3sXi.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.29968.png","used_models":0,"used_dl30":0},{"id":"2610.29357","date":"2026-10-22","title":"Memory-Centric Continual Learning for AI Agents","upvotes":43,"up_close":43,"up_d0":19,"up_d1":30,"comments":2,"peak":43,"purged":0,"closed":true,"early":false,"rank_day":17,"n_day":58,"github":null,"github_stars":null,"org":"SLAIAITP","org_fullname":"SLAI-AITP","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6a4c9bba52fc7b4faa5f9cf6/rAygogOXHG8he11FuN_If.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.29357.png","used_models":0,"used_dl30":0},{"id":"2610.30128","date":"2026-10-22","title":"Think Harder Where It Matters: Item-Level Adaptive Recursion for Sequential Recommendation","upvotes":41,"up_close":41,"up_d0":11,"up_d1":19,"comments":2,"peak":41,"purged":0,"closed":true,"early":false,"rank_day":18,"n_day":58,"github":"https://github.com/Elpmis117/LoopRec","github_stars":1,"org":null,"org_fullname":null,"org_avatar":null,"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.30128.png","used_models":0,"used_dl30":0},{"id":"2610.28711","date":"2026-10-22","title":"Qwen-Audio-3.1-ASR Technical Report","upvotes":37,"up_close":37,"up_d0":12,"up_d1":21,"comments":1,"peak":37,"purged":0,"closed":true,"early":false,"rank_day":19,"n_day":58,"github":null,"github_stars":null,"org":null,"org_fullname":null,"org_avatar":null,"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.28711.png","used_models":0,"used_dl30":0},{"id":"2610.13177","date":"2026-10-22","title":"CompactSplat: Spatially Adaptive Gaussian Distribution for Feedforward Scene Reconstruction","upvotes":34,"up_close":34,"up_d0":12,"up_d1":24,"comments":2,"peak":34,"purged":0,"closed":true,"early":false,"rank_day":20,"n_day":58,"github":"https://github.com/hebing-sjtu/CompactSplat","github_stars":3,"org":"SJTU","org_fullname":"Shanghai Jiao Tong University","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/1676013394657-63e5ee22b6a40bf941da0928.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.13177.png","used_models":0,"used_dl30":0},{"id":"2610.30120","date":"2026-10-22","title":"ALPS: Visual State Encoders for World Models","upvotes":34,"up_close":34,"up_d0":18,"up_d1":26,"comments":2,"peak":34,"purged":0,"closed":true,"early":false,"rank_day":20,"n_day":58,"github":"https://github.com/UCSC-VLAA/ALPS","github_stars":29,"org":"UCSC-VLAA","org_fullname":"UCSC-VLAA","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/645eb61da3c5cd8a16efffff/E7m3g_fFhz32pGsnK0eqX.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.30120.png","used_models":0,"used_dl30":0},{"id":"2610.19376","date":"2026-10-22","title":"DeLoop: Repeated Actions as Self-Supervision for Long-Horizon Agents","upvotes":33,"up_close":33,"up_d0":17,"up_d1":24,"comments":2,"peak":33,"purged":0,"closed":true,"early":false,"rank_day":22,"n_day":58,"github":"https://github.com/cskyl/deloop","github_stars":3,"org":"google","org_fullname":"Google","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/5dd96eb166059660ed1ee413/WtA3YYitedOr9n02eHfJe.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.19376.png","used_models":0,"used_dl30":0},{"id":"2610.29458","date":"2026-10-22","title":"Harness Scaling for Generalizable Agentic Reinforcement Learning","upvotes":33,"up_close":33,"up_d0":15,"up_d1":22,"comments":1,"peak":33,"purged":0,"closed":true,"early":false,"rank_day":22,"n_day":58,"github":"https://github.com/RUCAIBox/Harness-Scaling","github_stars":3,"org":"RUC","org_fullname":"Renmin University of China","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/61ac8f8a00d01045fca0ad2f/670IAX9A2-BflqA5MiSBW.jpeg","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.29458.png","used_models":0,"used_dl30":0},{"id":"2610.29560","date":"2026-10-22","title":"From VLM Planners to Robotic Agents: Self-Evolving Harnesses for History-Dependent Manipulation","upvotes":33,"up_close":33,"up_d0":10,"up_d1":23,"comments":1,"peak":33,"purged":0,"closed":true,"early":false,"rank_day":22,"n_day":58,"github":"https://github.com/StevenKKXS/SERA","github_stars":6,"org":null,"org_fullname":null,"org_avatar":null,"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.29560.png","used_models":0,"used_dl30":0},{"id":"2610.28679","date":"2026-10-22","title":"VerifierBench: Benchmarking the Generation of Executable Verifiers for LLM Agents","upvotes":32,"up_close":32,"up_d0":4,"up_d1":13,"comments":2,"peak":32,"purged":0,"closed":true,"early":false,"rank_day":25,"n_day":58,"github":"https://github.com/DwanZhang-AI/verifier-suite","github_stars":5,"org":null,"org_fullname":null,"org_avatar":null,"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.28679.png","used_models":0,"used_dl30":0},{"id":"2610.28763","date":"2026-10-22","title":"Automating ML for Science: Can Frontier Agents Climb Scientific Hills in the Wild?","upvotes":29,"up_close":29,"up_d0":13,"up_d1":20,"comments":2,"peak":29,"purged":0,"closed":true,"early":false,"rank_day":26,"n_day":58,"github":"https://github.com/maszhongming/agent4sci","github_stars":70,"org":"Anthropic","org_fullname":"Anthropic","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/1670531762351-6200d0a443eb0913fa2df7cc.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.28763.png","used_models":0,"used_dl30":0},{"id":"2610.30023","date":"2026-10-22","title":"RealMe: Immersive Volumetric Video from Minimal Captures","upvotes":28,"up_close":28,"up_d0":12,"up_d1":19,"comments":2,"peak":28,"purged":0,"closed":true,"early":false,"rank_day":27,"n_day":58,"github":"https://github.com/zju3dv/RealMe","github_stars":38,"org":"zju3dv","org_fullname":"zju3dv","org_avatar":"https://www.gravatar.com/avatar/38d28859223ed13a997edb0e9911614a?d=retro&size=100","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.30023.png","used_models":0,"used_dl30":0},{"id":"2610.27592","date":"2026-10-22","title":"Real In, Real Out: What If We Only Use Real Data for Scene Text Editing?","upvotes":27,"up_close":27,"up_d0":14,"up_d1":21,"comments":2,"peak":27,"purged":0,"closed":true,"early":false,"rank_day":28,"n_day":58,"github":"https://github.com/YesianRohn/TextRIRO","github_stars":1,"org":"Fudan-University","org_fullname":"Fudan University","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6437eca0819f3ab20d162e14/kWv0cGlAhAG3iNWVxowkJ.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.27592.png","used_models":0,"used_dl30":0},{"id":"2610.29441","date":"2026-10-22","title":"Sparsity Should Be ALLOTted by Output: Training-Free Allocation of Block-Sparse Attention for Video Diffusion Transformers","upvotes":25,"up_close":25,"up_d0":14,"up_d1":19,"comments":3,"peak":25,"purged":0,"closed":true,"early":false,"rank_day":29,"n_day":58,"github":null,"github_stars":null,"org":"KlingTeam","org_fullname":"KlingAI Research","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/60e272ca6c78a8c122b12127/ZQV1aKLUDPf2rUcxxAqj6.jpeg","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.29441.png","used_models":0,"used_dl30":0},{"id":"2610.30136","date":"2026-10-22","title":"Agentic World Action Models","upvotes":25,"up_close":25,"up_d0":11,"up_d1":17,"comments":1,"peak":25,"purged":0,"closed":true,"early":false,"rank_day":29,"n_day":58,"github":null,"github_stars":null,"org":null,"org_fullname":null,"org_avatar":null,"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.30136.png","used_models":0,"used_dl30":0},{"id":"2610.29545","date":"2026-10-22","title":"PACE: Periodic and Channel-Aligned Encoding of Four-Bit Recurrent States","upvotes":24,"up_close":24,"up_d0":19,"up_d1":20,"comments":7,"peak":24,"purged":0,"closed":true,"early":false,"rank_day":31,"n_day":58,"github":null,"github_stars":null,"org":"huberlin","org_fullname":"Humboldt Universität zu Berlin","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/-77OShpGOa1MGu9k5Oiwp.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.29545.png","used_models":0,"used_dl30":0},{"id":"2610.30108","date":"2026-10-22","title":"Multimodal Embodied Agents: A Unified Capability-centric Perspective from Computer-Use to Robot-Use","upvotes":24,"up_close":24,"up_d0":18,"up_d1":21,"comments":2,"peak":24,"purged":0,"closed":true,"early":false,"rank_day":31,"n_day":58,"github":"https://github.com/showlab/Awesome-Multimodal-Embodied-Agent","github_stars":206,"org":"showlab","org_fullname":"Show Lab","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/1671779505215-63a55320ce5763e06f78519c.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.30108.png","used_models":0,"used_dl30":0},{"id":"2610.21204","date":"2026-10-22","title":"Entertain, then Cancel: How Thinking Makes Reasoning Models Read Some and Or Literally","upvotes":23,"up_close":23,"up_d0":12,"up_d1":17,"comments":2,"peak":23,"purged":0,"closed":true,"early":false,"rank_day":33,"n_day":58,"github":"https://github.com/kouzhizhuo/entertain-then-cancel","github_stars":1,"org":"hongkongust","org_fullname":"The Hong Kong University of Science and Technology","org_avatar":"https://www.gravatar.com/avatar/4722e71b09c8244cf76e358091504d3b?d=retro&size=100","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.21204.png","used_models":0,"used_dl30":0},{"id":"2610.23344","date":"2026-10-22","title":"Language Models Learn to Improve Language Models","upvotes":23,"up_close":23,"up_d0":5,"up_d1":7,"comments":4,"peak":23,"purged":0,"closed":true,"early":false,"rank_day":33,"n_day":58,"github":null,"github_stars":null,"org":null,"org_fullname":null,"org_avatar":null,"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.23344.png","used_models":0,"used_dl30":0},{"id":"2610.27227","date":"2026-10-22","title":"Which Agent's Skill Should Change? Group-Relative Credit Assignment for Evolving Multi-Agent Teams","upvotes":22,"up_close":22,"up_d0":10,"up_d1":16,"comments":2,"peak":22,"purged":0,"closed":true,"early":false,"rank_day":35,"n_day":58,"github":"https://github.com/yananma1029/GRSE-Skills","github_stars":3,"org":"Mayoclinic","org_fullname":"Mayo Clinic","org_avatar":"https://www.gravatar.com/avatar/4b31495ac4e193847afde8644e1d8362?d=retro&size=100","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.27227.png","used_models":0,"used_dl30":0},{"id":"2610.28827","date":"2026-10-22","title":"WorldStream: Streaming Interactive 3D Worlds with Hierarchical Planning and Persistent State","upvotes":20,"up_close":20,"up_d0":9,"up_d1":13,"comments":1,"peak":20,"purged":0,"closed":true,"early":false,"rank_day":36,"n_day":58,"github":"https://github.com/krennic999/WorldStream","github_stars":23,"org":null,"org_fullname":null,"org_avatar":null,"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.28827.png","used_models":0,"used_dl30":0},{"id":"2610.29317","date":"2026-10-22","title":"SpatialBrain: Spatial Intelligence Growing from Experience","upvotes":20,"up_close":20,"up_d0":12,"up_d1":16,"comments":2,"peak":20,"purged":0,"closed":true,"early":false,"rank_day":36,"n_day":58,"github":null,"github_stars":null,"org":"SJTU","org_fullname":"Shanghai Jiao Tong University","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/1676013394657-63e5ee22b6a40bf941da0928.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.29317.png","used_models":0,"used_dl30":0},{"id":"2610.29784","date":"2026-10-22","title":"Evolving with Open Worlds: Towards Recursive Self-Improvement for Future Forecasting","upvotes":19,"up_close":19,"up_d0":11,"up_d1":15,"comments":2,"peak":19,"purged":0,"closed":true,"early":false,"rank_day":38,"n_day":58,"github":"https://github.com/RUCAIBox/DeepEvolve","github_stars":6,"org":null,"org_fullname":null,"org_avatar":null,"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.29784.png","used_models":0,"used_dl30":0},{"id":"2610.09861","date":"2026-10-22","title":"Toward Part-Aware Choral Transcription with singing voice assignment","upvotes":18,"up_close":18,"up_d0":8,"up_d1":12,"comments":2,"peak":18,"purged":0,"closed":true,"early":false,"rank_day":39,"n_day":58,"github":"https://github.com/Multimodal-Music-Research-Lab/Paw_Choral_AMT","github_stars":0,"org":"unswnkg","org_fullname":"UNSW","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/67c7bee047c2aa1353242e11/Ow-d_SWRBY3LU_elTqs2y.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.09861.png","used_models":0,"used_dl30":0},{"id":"2610.28905","date":"2026-10-22","title":"Hidden Messages Between Diffusion Language Model Agents: Causal Evidence for Latent Communication","upvotes":18,"up_close":18,"up_d0":7,"up_d1":11,"comments":1,"peak":18,"purged":0,"closed":true,"early":false,"rank_day":39,"n_day":58,"github":"https://github.com/alanrbtx/hidden-messages-diffusion-agents","github_stars":2,"org":"ITMO","org_fullname":"ITMO University","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/fBvTjWt5kZM802VUpqCno.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.28905.png","used_models":0,"used_dl30":0},{"id":"2610.27040","date":"2026-10-22","title":"ACE-Brain-0.6: Scaling Spatial Grounding for Physical Foundation Models","upvotes":17,"up_close":17,"up_d0":9,"up_d1":12,"comments":2,"peak":17,"purged":0,"closed":true,"early":false,"rank_day":41,"n_day":58,"github":"https://github.com/DAXIAORobotics/ACE-Brain-0.6","github_stars":8,"org":"ACERobotics","org_fullname":"ACE Robotics","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6a4a45de4ba0a921a91a548a/MSX_HJLXaBYKy2ZByYTnN.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.27040.png","used_models":0,"used_dl30":0},{"id":"2610.28940","date":"2026-10-22","title":"StyleStream 2.0: Fast and Controllable Streaming Voice Style Conversion","upvotes":17,"up_close":17,"up_d0":9,"up_d1":10,"comments":2,"peak":17,"purged":0,"closed":true,"early":false,"rank_day":41,"n_day":58,"github":"https://github.com/Berkeley-Speech-Group/StyleStream2.0","github_stars":5,"org":"Berkeley","org_fullname":"UC Berkeley","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/61ac8f8a00d01045fca0ad2f/0FjsTg2txEZZ4dEgmMnQL.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.28940.png","used_models":0,"used_dl30":0},{"id":"2610.28474","date":"2026-10-22","title":"Answering in the User's Variety: Calibrated Accommodation Pairs for Post-Training","upvotes":16,"up_close":16,"up_d0":8,"up_d1":13,"comments":2,"peak":16,"purged":0,"closed":true,"early":false,"rank_day":43,"n_day":58,"github":"https://github.com/kouzhizhuo/cap-accommodation","github_stars":0,"org":"hongkongust","org_fullname":"The Hong Kong University of Science and Technology","org_avatar":"https://www.gravatar.com/avatar/4722e71b09c8244cf76e358091504d3b?d=retro&size=100","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.28474.png","used_models":0,"used_dl30":0},{"id":"2610.29934","date":"2026-10-22","title":"ToM2Vec: Continuous Theory-of-Mind Tokens for Multi-Agent Coordination","upvotes":16,"up_close":16,"up_d0":7,"up_d1":8,"comments":2,"peak":16,"purged":0,"closed":true,"early":false,"rank_day":43,"n_day":58,"github":"https://github.com/nsivaku/tom2vec","github_stars":0,"org":"UNC-ChapelHill","org_fullname":"University of North Carolina at Chapel Hill","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/669f9c85bd649dba3b88e581/H5uB8_MCewnMtxEUnAvTL.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.29934.png","used_models":0,"used_dl30":0},{"id":"2610.29434","date":"2026-10-22","title":"RubatoCodec: Budgeted Adaptive Speech Tokenization with Streaming Decoding","upvotes":15,"up_close":15,"up_d0":8,"up_d1":11,"comments":2,"peak":15,"purged":0,"closed":true,"early":false,"rank_day":45,"n_day":58,"github":"https://github.com/third-intelligence/RubatoCodec","github_stars":13,"org":"third-intelligence","org_fullname":"Third Intelligence, Inc.","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/68e05b224c15b415781f3682/UnqR5Dai-ZdrlEAmRbISi.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.29434.png","used_models":0,"used_dl30":0},{"id":"2610.28027","date":"2026-10-22","title":"From RL Papers to Framework Code: Inclusion and Review in Seven Open-Source Training Frameworks","upvotes":14,"up_close":14,"up_d0":9,"up_d1":13,"comments":4,"peak":14,"purged":0,"closed":true,"early":false,"rank_day":46,"n_day":58,"github":"https://github.com/pth2002/rl-method-tracker","github_stars":2,"org":null,"org_fullname":null,"org_avatar":null,"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.28027.png","used_models":0,"used_dl30":0},{"id":"2610.28909","date":"2026-10-22","title":"DiffResearch: Accelerating Deep Research with Diffusion Language Models","upvotes":13,"up_close":13,"up_d0":6,"up_d1":9,"comments":1,"peak":13,"purged":0,"closed":true,"early":false,"rank_day":47,"n_day":58,"github":"https://github.com/alanrbtx/DiffResearch","github_stars":41,"org":"ITMO","org_fullname":"ITMO University","org_avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/fBvTjWt5kZM802VUpqCno.png","thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.28909.png","used_models":0,"used_dl30":0},{"id":"2610.30113","date":"2026-10-22","title":"Quant-Arena: Benchmarking LLM Agents for Quantitative Trading Research","upvotes":13,"up_close":13,"up_d0":4,"up_d1":6,"comments":1,"peak":13,"purged":0,"closed":true,"early":false,"rank_day":47,"n_day":58,"github":null,"github_stars":null,"org":null,"org_fullname":null,"org_avatar":null,"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.30113.png","used_models":0,"used_dl30":0},{"id":"2610.30090","date":"2026-10-22","title":"VX-One: Typed, Calibrated Decision Heads for Real-Time Perimeter Triage - An empirical study of unknown-attack honesty, cascade gray zones, and calibration under drift","upvotes":12,"up_close":12,"up_d0":9,"up_d1":10,"comments":2,"peak":12,"purged":0,"closed":true,"early":false,"rank_day":49,"n_day":58,"github":"https://github.com/vectrayx/vx-one","github_stars":0,"org":null,"org_fullname":null,"org_avatar":null,"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.30090.png","used_models":0,"used_dl30":0},{"id":"2610.30066","date":"2026-10-22","title":"Looped Uniform Diffusion Language Models","upvotes":11,"up_close":11,"up_d0":4,"up_d1":5,"comments":1,"peak":11,"purged":0,"closed":true,"early":false,"rank_day":50,"n_day":58,"github":null,"github_stars":null,"org":null,"org_fullname":null,"org_avatar":null,"thumbnail":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2610.30066.png","used_models":0,"used_dl30":0}],"hourly":{"ts":[1792641000,1792644600,1792648200,1792651800,1792655400,1792659000,1792662600,1792666200,1792669800,1792673400,1792677000,1792680600,1792684200,1792687800,1792691400,1792695000,1792698600,1792702200,1792705800,1792709400,1792713000,1792716600,1792720200,1792723800,1792727400,1792731000,1792734600,1792738200,1792741800,1792745400,1792749000,1792752600,1792756200,1792759800,1792763400,1792767000,1792770600,1792774200,1792777800,1792781400,1792785000,1792788600,1792792200,1792795800,1792799400,1792803000,1792806600,1792810200,1792813800,1792817400,1792821000,1792824600,1792828200,1792831800,1792835400,1792839000,1792842600,1792846200,1792849800,1792853400,1792857000,1792860600,1792864200,1792867800,1792871400,1792875000,1792878600,1792882200,1792885800,1792889400,1792893000,1792896600,1792900200,1792903800,1792907400,1792911000,1792914600,1792918200,1792921800,1792925400,1792929000,1792932600,1792936200,1792939800,1792943400,1792947000,1792950600,1792954200,1792957800,1792961400,1792965000,1792968600,1792972200,1792975800,1792979400,1792983000,1792986600,1792990200,1792993800,1792997400,1793001000,1793004600,1793008200,1793011800,1793015400,1793019000,1793022600,1793026200,1793029800,1793033400,1793037000,1793040600,1793044200,1793047800,1793051400,1793055000,1793058600,1793062200,1793065800,1793069400,1793073000,1793076600,1793080200,1793083800,1793087400,1793091000,1793094600,1793098200,1793101800,1793105400,1793109000,1793112600,1793116200,1793119800,1793123400,1793127000,1793130600,1793134200,1793137800,1793141400,1793145000,1793148600,1793152200,1793155800,1793159400,1793163000,1793166600,1793170200,1793173800,1793177400,1793181000,1793184600,1793188200,1793191800,1793195400,1793199000,1793202600,1793206200,1793209800,1793213400,1793217000,1793220600,1793224200,1793227800,1793231400,1793235000,1793238600,1793242200,1793245800,1793249400],"upvotes":[4,11,17,22,30,36,42,46,50,59,11,12,23,34,45,45,45,45,45,45,54,63,78,96,114,123,127,127,127,127,128,128,129,129,129,132,132,132,133,134,136,136,136,136,136,137,138,143,160,173,189,210,222,238,250,272,291,314,337,358,373,386,408,432,454,477,500,521,542,563,573,583,592,600,609,616,629,634,639,644,649,655,664,671,673,674,675,678,679,679,680,680,685,697,704,711,722,730,737,739,744,745,746,749,750,750,751,753,753,753,753,753,753,753,754,754,754,754,754,754,754,754,754,755,755,755,755,755,755,755,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757,757]},"context":null,"impact":null}