Search: “models”

I guarantee you are sleeping on small models. Deepseek V4 Flash can do ~80% of the tasks you ask Claude or Codex for. It is 137x cheaper per…

model-releasesclem-delangue--x

DeepSeek V4 Flash is a small language model capable of handling approximately 80% of common tasks performed by larger models like Claude or Codex, while offering significantly lower costs (137x cheape

Model Releases

IndicGuard: A Multilingual Safety Guard Model and Dataset for Indic Languages

DGX agent

arXiv:2606.22841v1 Announce Type: cross Abstract: As Large Language Models (LLMs) achieve widespread integration across diverse linguistic landscapes, ensuring their safety and alignment with regional

model-releasesarxiv-cs-lg

23 Jun 2026

Agents

Model council as a tool to push performance definitely works, I think the interesting next frontier is trying to scale this at the agent lev…

Model council as a tool to push performance definitely works, I think the interesting next frontier is trying to scale this at the agent level have been thinking a bunch about model routing and relate

agentsharrison-chase--x

MOOZY: A Patient-First Foundation Model for Computational Pathology

model-releasesarxiv-cs-cv

arXiv:2603.27048v3 Announce Type: replace Abstract: Computational pathology needs whole-slide image (WSI) foundation models that transfer across diverse clinical tasks, yet current approaches remain l

One-Step Flow Matching for Generative Modeling of Path-Dependent Physical Fields

arXiv:2606.22752v1 Announce Type: new Abstract: Physical simulations for intricate geometries with path-dependent constitutive models face difficulties due to the enormous computational cost they requ

researcharxiv-cs-lg

Tutorials

Provable Benefits of RLVR over SFT for Reasoning Models: Learning to Backtrack Efficiently

arXiv:2606.22938v1 Announce Type: new Abstract: Recent advances in large language models (LLMs) have demonstrated that reinforcement fine-tuning of pretrained base models can lead to significant gains

tutorialsarxiv-cs-lg

Training-free Task Classification for Multi-Task Model Merging

arXiv:2606.22589v1 Announce Type: new Abstract: Ever since the advent of foundation models and the pre-training-finetuning paradigm, there have been numerous efforts to merge multiple task-specific ex

it is indeed quite good! don't try it in claude code/codex - those harnesses are overly tuned for their proprietary models dcode (deepagents…

model-releasesharrison-chase--x

it is indeed quite good! don't try it in claude code/codex - those harnesses are overly tuned for their proprietary models dcode (deepagents code) is a model agnostic harness - try it there with @Fire

19 Jun 2026

Safety

A theory of learning data statistics in diffusion models, from easy to hard

arXiv:2603.12901v2 Announce Type: replace-cross Abstract: While diffusion models have emerged as a powerful class of generative models, their learning dynamics remain poorly understood. We address thi

safetyarxiv-cs-lg

Calibration Drift Under Reasoning: How Chain-of-Thought Budgets Induce Overconfidence in Large Language Models

arXiv:2606.11211v1 Announce Type: cross Abstract: The ability of large language models (LLMs) to express calibrated uncertainty is important for safe deployment. Chain-of-thought (CoT) reasoning is wi

Embodied-R1.5: Evolving Physical Intelligence via Embodied Foundation Models

arXiv:2606.11324v1 Announce Type: cross Abstract: We introduce Embodied-R1.5, a unified Embodied Foundation Model (EFM) that integrates comprehensive embodied reasoning capabilities, spanning embodied

Local Ai

Making Foresight Actionable: Repurposing Representation Alignment in World Action Models

arXiv:2606.12217v1 Announce Type: cross Abstract: World Action Models (WAMs) offer a promising route for robot manipulation by using video generation models to model future scene evolution before prod

local-aiarxiv-cs-ai

Physically Constrained Ensemble Gaussian Process Modelling for Expensive Quantum Systems with Heteroskedastic Noise

arXiv:2606.11240v1 Announce Type: cross Abstract: Accurate modeling of quantum many-body systems often requires computationally expensive simulations such as Density Matrix Renormalization Group (DMRG

Safety

To Intervene or Not: Guiding Inference-time Alignment with Probabilistic Model Blending

arXiv:2606.11201v1 Announce Type: cross Abstract: The wide deployment of LLMs has made model alignment necessary to make newly trained models safely and effectively respond to user instructions. Among

safetyarxiv-cs-ai

CableRobotGraphSim: A Graph Neural Network for Modeling Partially Observable Cable-Driven Robot Dynamics

model-releasesarxiv-cs-ro

arXiv:2602.21331v2 Announce Type: replace Abstract: General-purpose simulators have accelerated the development of robots. Traditional simulators based on first-principles, however, typically require

Data assimilation for subsurface flow using latent diffusion model parameterization: performance of ensemble-Kalman and Monte Carlo techniques

arXiv:2606.11140v1 Announce Type: cross Abstract: Data assimilation (DA) in subsurface flow entails calibrating model parameters to match observed data, typically at wells, while preserving geological

researcharxiv-cs-ai

Do Vision-Language Models See or Guess? Measuring and Reducing Textual-Prior Reliance with a Phrasing-Controlled Benchmark

model-releasesarxiv-cs-cl

arXiv:2606.10400v1 Announce Type: new Abstract: Vision-language models (VLMs) are increasingly deployed where answers must follow from what is in the image, yet they often answer from textual priors,

FailureScope: Cross-Regime Behavioral Diagnosis of Language Model Weaknesses

arXiv:2606.09878v1 Announce Type: new Abstract: Standard benchmarks report aggregate accuracy, but practitioners need to know which specific capabilities a model lacks. We introduce FailureScope, a be

Rotate2Think: Geometric Priming via Orthogonal Rotation to Improve Language Model Reasoning

arXiv:2606.09873v1 Announce Type: cross Abstract: Reasoning models achieve strong performance on challenging tasks by generating explicit intermediate reasoning traces before producing a final answer.

What Really Matters for Table LLMs? A Meta-Evaluation of Model and Data Effects

arXiv:2501.14717v2 Announce Type: replace Abstract: Table modeling has progressed for decades. In this work, we revisit this trajectory and highlight emerging challenges in the LLM era, particularly t

researcharxiv-cs-cl

A Dataset for Dynamic Human Preferences for Vision Language Models

arXiv:2606.07653v1 Announce Type: cross Abstract: Given the increased adoption of Vision Language Models (VLMs) in human-interactive settings, it is important that we evaluate how well these models ca

A retrieval conditioned rebinding circuit for dynamic entity tracking in large language models

arXiv:2606.08644v1 Announce Type: cross Abstract: To interpret context correctly and retrieve relevant information, large language models must bind entities to their attributes and update these bindin

Anthropic releases its first Mythos-class model Claude Fable

model-releasesthe-verge-ai

Anthropic just announced Claude Fable 5, a new AI model it said is the most powerful model it has ever made widely available. According to the company, Fable 5 'shows exceptional performance in softwa

Evaluating Hallucinations in Domain-Adapted Large Language Models

arXiv:2606.07521v1 Announce Type: cross Abstract: This study investigates the phenomenon of hallucinations in domain-adapted Large Language Models (LLMs), focusing on the fine-tuning of the Llama-2 mo

Fable 5 is the biggest step up I’ve felt in our models since Opus 4.5 back in November. After 4.5 came out I uninstalled my IDE when I reali…

model-releasesboris-cherny--x

Fable 5 is the biggest step up I’ve felt in our models since Opus 4.5 back in November. After 4.5 came out I uninstalled my IDE when I realized that I’d been doing 100% of my coding in a terminal for

Finite Certificates for In-Context Determinacy and a Threshold Theory of Emergence in Language Models

arXiv:2606.07623v1 Announce Type: new Abstract: This paper develops a model-theoretic framework for verifying context-conditioned language-model behavior by replacing benchmark labels with finite sema

MatSciBench: Benchmarking the Reasoning Ability of Large Language Models in Materials Science

arXiv:2510.12171v2 Announce Type: replace Abstract: Large Language Models have shown strong scientific reasoning ability, but their performance on materials science problems remains less studied. To f

Principled Agent Debate: Adversarial Arbitration for Sycophancy Reduction in Large Language Models

arXiv:2606.07532v1 Announce Type: cross Abstract: RLHF-trained models are systematically biased toward agreement over accuracy, a structural property of the training process. We present Principled Age

Stress-testing medical large language models reveals latent safety pathology beyond benchmark accuracy

arXiv:2606.07929v1 Announce Type: new Abstract: Large language models (LLMs) are entering clinical practice based on benchmark accuracy that may fail to detect safety-relevant failure modes. Here we p

Systematic LLM Translation of Legacy Scientific Code to Differentiable Frameworks: Application to a Land Surface Model

arXiv:2606.07681v1 Announce Type: cross Abstract: Differentiable programming offers transformative capabilities for scientific modeling, enabling gradient-based parameter estimation, sensitivity analy

Audio-Visual World Models: Grounding Multisensory Imagination for Embodied Agents

model-releasesarxiv-cs-cv

arXiv:2512.00883v3 Announce Type: replace-cross Abstract: World models simulate environmental dynamics to enable agents to plan and reason about future states. While existing approaches have primarily

Endogenous Resistance to Activation Steering in Language Models

arXiv:2602.06941v2 Announce Type: replace-cross Abstract: Large language models can recover mid-generation from task-misaligned activation steering, producing explicit verbal restarts (e.g., ``wait, t

Evidence Graph Consistency in Retrieval-Augmented Generation: A Model-Dependent Analysis of Hallucination Detection

arXiv:2606.06748v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) reduces but does not eliminate hallucination in large language models. Existing detection methods rely on flat si

Explain Like I'm 5 or Whatever I Choose: Evaluating the Interactive Potential of Language Model Responses

model-releasesarxiv-cs-cl

arXiv:2606.06788v1 Announce Type: new Abstract: Evaluations of large language models (LLMs) in scientific information seeking tasks have become increasingly use-centric, such as conducting live or mul

Last time around Apple released a lot of information about how their AI version of Siri worked between local and cloud models, not so much t…

model-releasesethan-mollick--x

Last time around Apple released a lot of information about how their AI version of Siri worked between local and cloud models, not so much this time It is nice to have a Gemma-like model on device, bu

Modular Monolingual Adaptation using Pretrained Language Models

arXiv:2606.06738v1 Announce Type: new Abstract: Building monolingual language models (LMs) for low-resource languages typically relies on adapting pretrained language models (PLMs) by finetuning the w

researcharxiv-cs-cl

PSA: Just added a few thousand chips, including B200s and B300s to our Dedicated Model Inference (http://api.together.ai/endpoints). With De…

model-releasestogether-ai--x

PSA: Just added a few thousand chips, including B200s and B300s to our Dedicated Model Inference (http://api.together.ai/endpoints). With Dedicated Model Inference, you can now on-click deploy our Bla

This is super big I think this is the first useful speculative decoding method deployed on a big quasi frontier model Massive unlock @fi5662…

model-releasesjeremy-howard--x

This is super big I think this is the first useful speculative decoding method deployed on a big quasi frontier model Massive unlock @fi56622380 🚀 1,000+ TOKENS/S ON A 1T MODEL! 🚀 We are thrilled to r

When Large Language Models Fail in Healthcare: Evaluating Sensitivity to Prompt Variations

arXiv:2606.07237v1 Announce Type: cross Abstract: Large Language Models (LLMs) are increasingly used in healthcare for tasks such as clinical question answering, diagnosis support, and report summariz

There was an inflection point recently where the tide shifted to model pickers and OSS Mix of tokenmaxxing/cost fatigue, nemotron coalition,…

model-releasesclem-delangue--x

There was an inflection point recently where the tide shifted to model pickers and OSS Mix of tokenmaxxing/cost fatigue, nemotron coalition, harness step functions, brains/claws, etc Long live those w

7 Jun 2026

No Need to Train Your RDB Foundation Model

arXiv:2602.13697v2 Announce Type: replace Abstract: Relational databases (RDBs) contain vast amounts of heterogeneous tabular information that can be exploited for predictive modeling purposes. But si

6 Jun 2026

Humans' ALMANAC: A Human Collaboration Dataset of Action-Level Mental Model Annotations for Agent Collaboration

model-releasesarxiv-cs-cl

arXiv:2606.06388v1 Announce Type: cross Abstract: Recent advances in LLM agents have enabled complex cognitive capabilities, such as multi-step reasoning, planning, and tool use, that increasingly pos

5 Jun 2026

Knowledge Distillation for Visual Autoregressive Models

arXiv:2606.06078v1 Announce Type: new Abstract: Autoregressive (AR) image generation models are highly expressive but computationally intensive, motivating effective model compression. Knowledge disti

researcharxiv-cs-cv

5 Jun 2026

ChessMimic: Per-Rating Transformer Models for Human Move, Clock, and Outcome Prediction in Online Blitz Chess

arXiv:2606.04473v1 Announce Type: cross Abstract: We present ChessMimic, a system of three small encoder-only transformers - for move, thinking-time, and outcome prediction - conditioned on the positi

Exposing Blindspots: Cultural Bias Evaluation in Generative Image Models

model-releasesarxiv-cs-cv

arXiv:2510.20042v3 Announce Type: replace Abstract: Generative image models produce striking visuals yet often misrepresent culture. Prior work has examined cultural bias mainly in text-to-image (T2I)

Flow Matching Calibration for Simulation-Based Inference under Model Misspecification

arXiv:2509.23385v5 Announce Type: replace-cross Abstract: Simulation-based inference (SBI) is transforming experimental sciences by enabling parameter estimation in complex non-linear models from simu

From Symbolic to Geometric: Enabling Spatial Reasoning in Large Language Models

arXiv:2606.04381v1 Announce Type: cross Abstract: Recent large language models (LLMs) often appear to exhibit spatial reasoning ability; however, this capability is largely symbolic, arising from patt

New course on serving LLMs efficiently -- how do you serve models to many concurrent users at low latency and reasonable cost? This short co…

model-releasesandrew-ng--x

New course on serving LLMs efficiently -- how do you serve models to many concurrent users at low latency and reasonable cost? This short course is built with @RedHat and taught by @cedricclyburn. Eff