Veille scientifique · Page 9

INSPIRE: An Internalize-Then-Improve Approach for Example-Driven Mathematical Reasoning

Mathematical reasoning · preprint

Mathematical reasoning has seen rapid progress in large language models (LLMs), yet existing methods optimize predominantly for final-answer correctness, raising the question whether models truly

ProofEvolve: Neuro-Symbolic Evolution for Formal Automated Theorem Proving

Formal verification · preprint

Automated theorem proving offers a natural foundation for recursive self-improvement in scientific discovery. However, existing neural provers do not fully preserve this recursive structure, where

Intelligent transcription with Gemini 3.5 Transcribe

Multimodal AI · provider_announcement

Now you can get more intelligent speech-to-text transcription with Gemini 3.5 Transcribe.

Quantization-Aware Healing: a compressed, 4-bit model that outperforms its full-precision original

Compute & efficiency · primary_technical_publication

Granite 4.2 LLMs: How They're Built

AI models · primary_technical_publication

Wire It, Run It, Deploy It: AI Workflows in Gradio

Agent systems · primary_technical_publication

FLARE: Verifying MILP Reformulations with LLM-Based Theorem Proving

Formal verification · preprint

Mixed-Integer Linear Programming (MILP) is a fundamental tool for combinatorial optimization with extensive real-world applications. A central challenge is designing computationally efficient MILP formulations. Large

More Accurate or More Efficient? Evaluating Locally Deployed Compact Open-Weight Language Models for Mathematical Reasoning

Mathematical reasoning · preprint

Large language models are increasingly deployed on local hardware for privacy, cost, and accessibility reasons. Yet many evaluations emphasize accuracy while fewer quantify local runtime

Measuring benchmark optimization in speech recognition

Evaluation & safety · primary_technical_publication

How Hugging Face Inference Endpoints, Jobs, and Buckets Power Search on Papers with Code

Compute & efficiency · primary_technical_publication

DirEAG: Dirichlet Evidence Aggregation for Calibrating Verbalized Confidence in Mathematical Reasoning

Mathematical reasoning · preprint

Reliable confidence estimation is essential for using large language models in mathematical reasoning, but black-box verbalized confidence is difficult to calibrate. When the same problem

From Atari to EVE Online: Building on 15 Years of AI Research in Games

AI models · provider_announcement

Google DeepMind partners with game studios to prototype breakthrough AI gameplay.

Up to 3.2x Faster Inference with LFM2.5-DSpark

Compute & efficiency · primary_technical_publication

SignalReasoner: Assessing the Upper Bound of 3B Models for Signal Mathematical Reasoning

Mathematical reasoning · preprint

Post-training with supervised chain-of-thought fine-tuning and reinforcement learning from verifiable rewards has substantially improved the mathematical reasoning capabilities of large language models (LLMs). However, their

How Much Memory Does Your Agent Actually Need?

Agent systems · primary_technical_publication

Introducing Gemini 3.7 Flash

AI models · provider_announcement

Record, train, and deploy from one place with Strands Agents, LeRobot, and Hugging Face Storage Buckets

Agent systems · primary_technical_publication

Putting sign language AI into users’ hands

AI models · provider_announcement

Introducing sign-language-to-text (SL2T), our breakthrough model powering new sign language features for Deaf and hard of hearing users.

Reward-Oracle MCTS for Formal Theorem Proving: Sample-Efficient Search and the Need for Kernel-Level Proof Auditing

Formal verification · preprint

Formal theorem proving with large language models remains challenging due to the difficulty of navigating large proof search spaces efficiently. Existing tree search approaches either

MathShikkha: A Controlled Study of Answer-Only and Chain-of-Thought Supervision for Bangla Mathematical Reasoning in Small Language Models

Mathematical reasoning · preprint

Mathematical reasoning remains challenging in low-resource languages such as Bangla. We study whether teacher-generated Bangla Chain-of-Thought (CoT) supervision provides benefits beyond ordinary supervised fine-tuning. We

WeatherNext: AI model achieves breakthrough in forecasting cyclones

AI models · provider_announcement

Escaping Confidence Trap: Evolutionary Decoding for Mathematical Reasoning in Diffusion LLMs

Mathematical reasoning · preprint

Diffusion large language models (dLLMs) have emerged as a promising alternative to autoregressive LLMs, offering efficient generation through block-wise progressive unmasking. However, their strong general-purpose

ReDiPPO: Reference-Guided Value Calibration and Discrepancy-Aware Token Reweighting for Mathematical Reasoning

Mathematical reasoning · preprint

Reinforcement learning has emerged as an effective paradigm for enhancing the mathematical reasoning capabilities of large language models. Among existing policy optimization methods, Proximal Policy

Gemini Robotics ER 2: powering robotics with video understanding, task orchestration, and multi-robot collaboration

Agent systems · provider_announcement

Gemini Robotics ER 2 helps robots reason, collaborate, and solve real-world tasks. It represents a step change in video understanding, tool orchestration, and multi-robot collaboration