PapersWithCode (@paperswithcodee) header
PapersWithCode (@paperswithcodee) profile picture

Trending AI research papers with code, datasets and leaderboards.

Global
Joined October 2026
Photo by PapersWithCode (@paperswithcodee)Photo by PapersWithCode (@paperswithcodee)Photo by PapersWithCode (@paperswithcodee)Photo by PapersWithCode (@paperswithcodee)Photo by PapersWithCode (@paperswithcodee)Photo by PapersWithCode (@paperswithcodee)Photo by PapersWithCode (@paperswithcodee)Photo by PapersWithCode (@paperswithcodee)Photo by PapersWithCode (@paperswithcodee)Photo by PapersWithCode (@paperswithcodee)
Replying to @Apple
@Apple and Cornell researchers, with Vladlen Koltun, just released FastFrames, autoregressive video generation fast enough for real-time interaction! 🎮 "FastFrames: Video Generation at Interactive Latency" This paper shows that steps usually treated as fixed implementation details, like VAE decoding and KV cache updates, account for over half of end-to-end inference latency. It covers lightweight replacements for both, a 1.3B model at 36 ms per frame on a single B200 matching a state-of-the-art model nearly 8x slower, a 14B model at 45 ms per frame with significantly better quality, camera- and physics-conditioned video generation, and much more! Read this excellent paper by Gene Chou, Vladlen Koltun and collaborators: paperswithcode.co/paper/2611…
92
Tencent's WeMM-Video-9B is now the best open model on DailyOmni: 83.6% accuracy, just 1 point behind closed Qwen3.5-Omni-Plus at 84.6%! 🎬 It's a 9B omni model (text, image, audio and video), with weights on Hugging Face under Apache 2.0. It comes from WeMM-Video, Tencent and Tsinghua's new family of omni-modal models from 2B to 35B-A3B, trained on about 1.2 trillion tokens.
1
99
Seyed Mahmoud Sajjadi Mohammadabadi just released a sobering meta-analysis for anyone building multi-agent LLM systems! 🤖 "More Agents, More Samples, or More Thinking? Untangling Coordination, Ensembling, and Test-Time Compute in Multi-Agent LLMs" This paper shows that multi-agent gains largely disappear once you compare against a single model given the same number of samples or the same compute budget. It covers 516 multi-agent configurations from nine evaluations, an average accuracy change of -0.042 against matched baselines, multi-agent setups at or below the baseline in 74% of comparable contrasts, when coordination can pay off (partitioned evidence, external verification, error diversity across models), and much more! Read this excellent paper by Seyed Mahmoud Sajjadi Mohammadabadi: paperswithcode.co/paper/2611…
1
192
Maksym Andriushchenko and collaborators at ELLIS Institute Tübingen just released a fascinating case study: an autonomous agent running its own open-model organisation on Hugging Face! 🤗 "Autonomous agents can run successful open-model organisations. Evidence from an 8-week case study" This paper shows what happens when you give an agent a blank Hugging Face account, compute, and one goal: earn real adoption. It covers Horizon, an agent built on Opus 5.5 running 24/7 in Claude Code, 600+ GPU-hours, 38 multilingual models fine-tuned from mmBERT (prompt-injection guards, PII redactors, rerankers), 50,000+ downloads, use in 100+ GitHub repos, and much more! Read this excellent paper by @AWuhrmann, Maksym Andriushchenko and collaborators: paperswithcode.co/paper/1334…
2
1
4
791
HKUST, CUHK and Zhejiang University researchers, with Ser-Nam Lim, just released WorldDraft, a sharp test of a big question for video generation: if code builds the world, can video models just render it? 🎬 "WorldDraft: If Code Builds the World, Can Video Models Just Render It?" This paper checks whether video models can keep prescribed 4D motion and camera paths while rendering the appearance a text prompt asks for. It covers 432 paired test cases across 48 executable 4D world families, 12 video models evaluated, why richer drafts anchor structure but hurt appearance fidelity, source leakage of up to 64% under conflicting cues, WorldDraft-Primer, a plug-and-play visual prompt that improves both, and much more! Read this excellent paper by Zhiyu Hou, Wen-Jie Shu and collaborators: paperswithcode.co/paper/2611…
1
204
From Sichuan University and UESTC (Weiwei Li, Wen Li and collaborators). It improves worst-group accuracy on seven benchmarks across vision, language, speech and medical imaging, and code and checkpoints are released. Full leaderboard: paperswithcode.co/benchmark/…
82
CHORD is the new SOTA on Waterbirds: 93.11% worst-group accuracy! 🐦 It just edges out HSFM (93.1%), with DaC-C (92.6%) next. Waterbirds crosses land and water birds with land and water backgrounds, so a model that leans on the background shortcut fails on the rare combinations. CHORD's idea: a bias hidden in one feature space may be clearer in another. It debiases a ResNet-50 through consensus across several representation spaces, with no bias labels.
1
137
The same recipe turns the four-slot Streaming Sortformer into an open-speaker system. Project page: kyudan1.github.io/eDIAR/
49
KAIST researchers, with Jaegul Choo, just released eDIAR, speaker diarization for hour-long meetings with no cap on how many people speak! 🎙️ "eDIAR: Training-Free Open-Speaker Streaming Diarization for Long Meetings and Large Archives via e-means" This paper removes the fixed speaker limit of slot-based diarizers like Streaming Sortformer and Nemotron-3, without any retraining. It covers e-means clustering inside a slot once every slot is used, bounded-window re-clustering so cost per update doesn't grow with the meeting, far better accuracy than Streaming Sortformer on hour-long meetings, archive-scale clustering several times faster than Lloyd's k-means, and much more! Read this excellent paper by Kyudan Jung, JiHwan Kim and collaborators: paperswithcode.co/paper/2611…
1
94
M3OPD-Reasoner-9B is the new SOTA on LogicVista: 79.6% accuracy! 🧠 That's 11.1 points ahead of the previous best, MMVistaReason-9B (68.5%). The recipe: train separate teachers for image STEM, perception and video, distill them into one Qwen3.5-9B student with on-policy distillation, then add RL on top. Data, training code and every checkpoint are released, with weights on Hugging Face under Apache 2.0.
1
66
Replying to @AntGroup
@AntGroup just released Inpriora, a pretrained tabular foundation model that learns your table in context, with no gradient updates at prediction time! 📋 "Inpriora: A Pretrained Tabular Foundation Model for Classification and Regression" This model handles both classification and regression, and was trained entirely on synthetic data. It covers adaptive Fourier features plus within-column ranks to encode each cell, a dual-branch in-context learner with separate attention projections for context and query rows, a 1831 Elo and 6th place among 79 entries on TabArena-v0.1, and much more! Read this excellent work from Ant Group's AI4Data team: paperswithcode.co/paper/1334…
1
1
140
Congrats @AIatMeta, KAUST and Princeton: AVTR-9B is the new SOTA on CharadesSTA video temporal grounding! 🎬 It reaches 64.7 mIoU, ahead of VideoAuto-R1 (63.7) and TempSamp-R1 (62.1). AVTR adds "visual imagination" to reasoning: instead of only thinking in text, the model can revisit what it saw through generative visual reflection, and learns when to use each. The weights are on Hugging Face.
1
81