Hugging Face Daily PapersThrough reinforcement learning (RL) and fine-grained rewards, researchers have developed an interface called BookNavigator to help large language models (LLMs) autonomously retrieve, learn, and apply knowledge from long reference materials to solve mathematical problems.
Hugging Face Daily PapersSteerDistill quantizes LLM agents through closed-loop teacher guidance, achieving an average score 1.67-2.18 points higher than the strongest quantization baseline on W4A16, a success rate of 1.84 percentage points on τ^2-Bench, and reducing training GPU hours by 67%.
Hugging Face Daily PapersTOME-Math transforms 2,118 university mathematics textbooks into approximately 380K structured proof supervision records, preserving chapter order, proposition-proof links, and dependency references, used for fine-tuning to improve the mathematical reasoning ability of language models by 16.7 percentage points, surpassing all three public instruction datasets. This corpus will be released to promote structured mathematical supervision research.
MIT News · AIMIT researchers have developed a training technique that significantly reduces data volume and improves the performance and adaptability of climate simulators (AI models), enabling them to quickly simulate the impact of human activities on future climate in seconds, such as predicting temperature or sea level changes, thereby more reliably testing thousands of climate scenarios and identifying unintended consequences of regional interventions. The algorithm generates optimized synthetic climate scenario data instead of relying on future emission scenarios based on real-world trajectories, significantly shortening production time and allowing the model to accurately predict scenarios not present in the training data.
Latent SpaceTypeSafe AI's Jev is a non-generative system-one model that can return structured choices, scores, or probabilities, faster and cheaper than inferential LLM (system-two models). Jev is suitable for zero-shot flexibility, requiring no repeated fine-tuning, and users can achieve reliable intelligent automation without needing to understand its internal model type. Jev can be integrated into code, enabling the construction of smarter software through rapid calls, with its application potential likened to the exploration stage of the early internet.
Hugging Face Daily Papers6565The One Table Benchmark tested the table generation, imputation, and prediction capabilities of 46 models on 56 datasets, finding that these tasks share conditional information but have different fidelity and target interactions. Fixed models exhibit a trade-off between utility and fidelity, while frozen base models perform outstandingly in prediction and numerical imputation, indicating that unified table models have the potential to improve classification reconstruction and joint sampling.
Hugging Face Daily PapersEmbodied First Exam has been released, including 35 sets, 47 robots, and 1,156 tasks, to evaluate the control capabilities of coding agents under physical world constraints, with GPT-6 Astra in Codex achieving a success rate of 34% in 456 core tasks under standard settings.
Intelligent agents such as Gemini, ChatGPT, Claude, and Grok discovered an officially unrecorded Mayan site in Mexico by analyzing NASA LiDAR data, completing the entire process from inspiration to government formal reply in less than 72 hours. The site is located on a mountain top in the jungle, containing structures such as plazas, pyramids, and courtyards. The agents identified and submitted reports to the National Institute of Anthropology and History of Mexico (INAH), which confirmed that the location had not been registered (later verified that the site had been mapped by anthropologists).
OpenRouter BlogRunning the same tasks on OpenRouter, GPT-6.1 Sol passed 180/180 tests in 60 synthetic tasks, DeepSeek had the lowest cost passing 157/180 times, but the correlation with the public ranking is low (Spearman correlation coefficient 0.109). Need to check the number of task repetitions, the number of answers, and billing verification.
AWS Machine Learning Blog5858Amazon SageMaker HyperPod detects silent data corruption (SDC) by adding three checks in the PyTorch DDP training loop, including non-finite gradient checks, loop replay audits, and weight fingerprint comparisons. When issues are detected, suspicious nodes are isolated and replaced by HyperPod, having been tested and verified on a 4 ml.p5.48xlarge instance cluster.
Databricks BlogDatabricks detects credential-driven attacks by analyzing audit logs and Genie, finding stolen credentials listed from new IPs in secret scopes, directories, and repositories within minutes, then reading secrets containing service principal credentials, and subsequently having the service principal create functions and join admin groups, these non-linear event sequences reveal intrusion behavior.
Hugging Face Daily Papers6565MACaQ is a post-training quantization method for MXFP4 W4A4, which reduces costs while maintaining accuracy in language and diffusion models by using online activation transformations and Hessian-aware refinement to independently quantize each linear layer. MACaQ achieves only a 2.3% end-to-end overhead increase on Qwen3-32B with vLLM, has a lower perplexity than DuQuant++ and MR-GPTQ, and reaches the best perplexity of Llama-3-8B in 11 times shorter quantization time.
AWS Machine Learning Blog1212AWS re:Invent 2026 AI/ML themed symposium will be held from November 30 to December 4, offering over 140 workshops covering full-stack content such as AI chips, model training and inference, and agent development management.
MarkTechPost6060Google DeepMind releases AlphaGenome 2, a DNA model that predicts the effects of specific cell type mutations. The model reads 1 Mb DNA and target cell state, significantly improving single-cell eQTL prediction performance through conditional sequence to functional architecture, increasing recall from 23% to 43% on the TenK10K benchmark, and can adapt to unseen datasets without fine-tuning.
Google Research5858Google Research releases a wearable device-driven CKM risk assessment framework, predicting future health trajectories through an explainable health age model (CHA). The framework is trained on wearable data from 2.2 million weeks using the SensorWM model, providing personalized 12-week counterfactual health simulations combined with language models to generate clinically validated guidance plans. The CHA model integrates modifiable indicators such as heart rate, steps, and sleep to predict disease risk over 68,000 person-years, improving the predictive performance of clinical indicators.
Hugging Face Daily PapersFINCOSMOS has introduced an open-world evaluation paradigm for evaluating intelligent agents in the financial field. In 150 simulated user interactions, no session was accepted because all sessions failed during program checks, with the agent reporting only 54-64% of the undisclosed details. Human experts attributed 88.33% of the failures to the agent.
Apple Machine Learning ResearchWeighted State Souping proposes query dependency weighting to combine cached State state encodings, significantly improving long-context performance on tasks such as single-hop and multi-hop document question answering, pinpoint retrieval, variable tracking, and tool use. The method can be adapted into an efficient sparse variant, achieving near-constant encoding costs while maintaining (or sometimes improving) performance.
Apple Machine Learning ResearchState Space Models (SSMs) lag behind Transformers in contextual learning and precise retrieval tasks, as their gating mechanism leads them to first learn memory solutions, delaying or preventing the convergence of correct contextual learning solutions, but the gating mechanism is beneficial for improving long sequence generalization.
Apple Machine Learning ResearchResearchers evaluated the performance of Large Language Model (LLM) judges in personalized writing tasks and found that the quality of judges is weakly correlated with general LLM capabilities measured by MMLU (Pearson’s ρ ≤ 0.2), and their performance is influenced by writing tasks and evaluation criteria (e.g., general helpfulness vs. stylistic personalization). LLM evaluators were more consistent and performed better in pairwise comparisons than in independent scoring responses. The 'strongest' LLM did not reliably replicate human evaluations, possibly due to being overly sensitive to subtle details overlooked by human annotators and weaker LLMs.
NVIDIA Technical Blog6565Using NVIDIA DSX AI Factory, you can track actual AI workloads running on NVIDIA Dynamo, Kubeflow, or Slurm platforms, exceeding the visibility provided by platform configuration; application teams independently define and update workloads, and deploying a package cannot determine the workload components running on a node.
Hugging Face Blog3838Hugging Face's blog introduces the Token Pooling perception training method for multi-vector embedding models, which compresses storage by merging similar tokens through HierarchicalTokenPooling.
Simon WillisonAnthropic disclosed in its IPO roadshow video that the Claude model now supports four major chip platforms: AWS Trainium, Google TPUs, NVIDIA GPU, and AMD GPU, a result of its early multi-chip strategy that reduced the deployment time for new platforms from months to three days.
Ars Technica · AIFeaturedSame story · 2 outlets7575Google DeepMind researchers used AI agents to formally verify multiple key cryptographic guarantees of the Signal protocol, completing the project as a proof of concept within four weeks. This work provides independent verification for encryption used by platforms such as Signal, Google Messages, and WhatsApp, while also demonstrating the formal verification capabilities of the new AI framework AlphaProof Code.
IBM Research Blog3838IBM Quantum Starling will be delivered in 2029, capable of performing 100 million hard operations on 200 qubits. IBM has validated the c-coupler technology, achieving dense non-local qubit connectivity, and has improved the qLDPC code to require only five qubit connections instead of six.
Hugging Face Daily PapersTree-Ising fixes the inference loop of large language models by solving sparse Ising/QUBO problems on BF16 parameters, improving mathematical benchmark accuracy with fewer weight edits without gradient or LLM-token costs.
vLLM Blog5858vLLM extends multi-modal service capabilities through the EPD (Encoder/Prefill/Decode) separation technology, supporting independently deployed visual encoders. vLLM v0.32.0 includes MiMo and XPU support, and the metadata-only request feature of DeepSeek V4.1 Flash has been open-sourced.
Hugging Face Daily Papers"WikiTrace Links Each Distillation Experience to Its Source Trajectory and Reasoning Chain, Allowing Large Language Model Agents to Verify and Correct Memory Before Action, Improving Performance and Reducing Degradation Caused by Memory in Five Intelligent Agent Benchmark Tests. The System Stores Each Experience as a Condensed Insight and Links It to a Summary of the Reasons for Its Success or Failure and the Original Trajectory, Enabling Agents to Supplement Missing Details or Correct Insights When Executing New Tasks Through Linking Tools. The Results Show It Outperforms Existing Memory Methods and Baselines Using the Same Insights"
Google AI BlogGoogle Arts & Culture open-sources multimodal model Veo, transforming the early memories of artist Esther Mahlangu into video, completing an online retrospective exhibition of her life. The model is based on 45 stories and archival materials, generating images and converting them into video to present key memories such as her 10th birthday discovery of her painting talent.
Ant Group inclusionAI · Hugging Face new modelsinclusionAI has released the ZoomGain-4B large language model, achieving a score of 61.08 on ZoomGainBench and performing excellently on multiple fine-grained perception and general perception benchmark tests; the model is built with supervision based on the global-to-local visual perception gap mined from global image details, supporting English and Chinese Q&A, and can be downloaded via Hugging Face.
Hugging Face Daily PapersInfraRSI transforms open-ended human instructions into validated training optimization through a five-stage workflow, achieving up to 2.75x acceleration in four single-node and multi-node dense as well as Mixture-of-Experts (MoE) LLM workloads, and increasing the average model FLOPs utilization (MFU) from 17.4% to 30.0%, surpassing existing tuners and coding agents.
Hugging Face Daily Papers6565SWAM extends the World Action Model through cross-modal alignment, trained on 47,000 hours of multimodal data, blending egocentric video and UMI demonstrations to enhance control generalization, ultimately achieving a 40.00% success rate on three real-world tasks, surpassing the baseline of 23.33%.
Hugging Face Daily PapersWorld In-Context Learning (WICL) method freezes the model, practices tasks in simulators, and writes feedback into the context rather than weights, enabling robots to learn operational skills from a single demonstration. RoboScript implements WICL, achieving an 83.4% success rate on 50 tasks in RoboTwin 2.0 with no gradient updates from a single demonstration, surpassing the strongest training strategy by over 20 points, and supports cross-visual distribution generalization, improvement through reflection, base model extension, and cross-platform memory migration.
Hugging Face Daily Papers6565The DexForesight framework implicitly learns spatial foresight by injecting a dynamic perception world representation into the latent space of dexterous strategies, increasing the success rate of basic strategies from 49.4% to 60.5% on 11 DexJoCo simulation tasks and achieving a 58.3% success rate on 3 real robotic dexterous operation tasks.
Hugging Face Daily Papers6565MMDM evaluated 28 decision models, showing that most models perform well on common knowledge but poorly on interdependent options, constrained decision-making, and multiple-choice tasks. For example, Perplexity Decider-V1.2 achieved 78.75% accuracy on single-choice knowledge, but only 53.95% and 47.95% respectively on option logic and constrained planning, while the best accuracy of all models on hard multiple-choice was only 36.81%.
Hugging Face Daily Papers6565Infrathon conducted 150 long-term AI infrastructure optimization task evaluations on six models, consuming a total of 10,000 GPU hours, and found that the model rankings changed with the evaluation duration. 135 tasks achieved further optimization in the latter half of the budget, and Claude Opus 5.5 increased the KV cache compression rate from 6.82% to 19.86%, exceeding human design baselines. Analysis shows that task violations, unreliable testing, and premature completion are the main obstacles to continuous optimization.
Modern Web Guidance embeds web platform expertise, best practices, and browser compatibility data into code agents, helping them avoid outdated patterns and shift towards leveraging modern web platform capabilities.
Hugging Face Daily PapersdNaViT is a discrete native resolution vision Transformer that matches or exceeds specific baselines in visual understanding, generation, and editing tasks through grouped product quantization and dual-branch targets. It performs excellently on 12 visual understanding benchmarks, 4 generation benchmarks, and one image editing benchmark, proving that discrete vocabulary can be a universal visual representation
