DailyPapers

Paper: paperswithcode.co/paper/2610… Project: tome-math.github.io/ Dataset coming soon on Hugging Face.

TOME-Math: A structured textbook corpus for math reasoning Curated from 2,118 university textbooks into 380K proof-supervision records, preserving chapter order and proposition-proof links. Fine-tuning on it lifts average math benchmark performance by 16.7 percentage points.

Check it out: huggingface.co/nvidia/Kimi-K… Ready for vLLM and SGLang, optimized for Blackwell GPUs with NVIDIA Model Optimizer.

NVIDIA just released a 4-bit quantized Kimi-K3.1-Pro on Hugging Face 2.8T parameters, 104B active, near-lossless.

Try it here: huggingface.co/facebook/meta…

Meta just released MetaEncoder on Hugging Face A unified multimodal encoder for text, images and video, with SOTA retrieval and decision support.

Paper: paperswithcode.co/paper/2611… Dataset: huggingface.co/datasets/Open…

GTAGym by Alibaba's OpenMOSS team A real AAA game tasks environment for evaluating and training agents on GTA V, with 150 tasks across 11 categories and RL support.

Paper: paperswithcode.co/paper/2611… Project: yukyeongleee.github.io/wikit… Code: github.com/yukyeongleee/wiki…

Don't just trust your memory WikiTrace links each lesson to its original trajectory, so agents can check the evidence before acting. It reduces memory-induced failures by up to 9.7% across five benchmarks.

Get it here: huggingface.co/datasets/face… Paper: arxiv.org/abs/2610.35902

Meta just released XRStereo on Hugging Face A synthetic stereo video dataset with ~18k sequences - rectified stereo pairs, dense disparity, poses & intrinsics Built to train StereoSplat for metric-scale novel view synthesis

All code is open-source: github.com/ArcheOne-AI/RoboS… Paper: paperswithcode.co/paper/2611…

Learning from the world, not from weights RoboScript lets a frozen model learn manipulation from one demonstration by rehearsing in a simulator, writing the feedback into its context as a Task Memory — no gradient updates, and it hits 83.4% on 50 RoboTwin 2.0 tasks.

Paper: paperswithcode.co/paper/2611… It boosts Qwen3.5-2B on GSM8K from 21.6% to 69.4% — no gradients, no agent loops.

Tree-Ising: repairing LLMs with bit-level Ising edits This method fixes model weights by solving sparse discrete optimization problems on bit patterns, outperforming frontier-agent weight editing with just 4 forward passes and zero extra tokens.

Paper: paperswithcode.co/paper/2611… Dataset: huggingface.co/datasets/shen… Leaderboard: mmdm-benchmark.github.io Explore how 28 decision models perform across realistic tasks like agent tracking and rubric scoring.

MMDM: Measuring Massive Multitask Decision Making A new benchmark evaluates System-1 decision models with 5,448 questions across 16 categories, testing single-select, multi-select, and soft-label choices under long contexts and up to 1,024 options.

Paper: paperswithcode.co/paper/2611… Model: huggingface.co/meituan-longc… Project: lixiaotong97.github.io/dNaVi…

dNaViT: discrete codes as universal visual representation Meituan's new tokenizer maps native-resolution images to a shared discrete vocabulary, delivering competitive performance across understanding, reconstruction, generation, and editing – all in one model. Code and weights on Hugging Face.