model releases

Sovereign AI for all.

Cohere advocates for democratizing access to sovereign AI systems, enabling organizations and nations to develop and deploy their own AI models independently rather than relying on centralized provide

model-releasescohere--x

Sparse Autoencoders Reveal Interpretable and Steerable Features in VLA Models

arXiv:2603.19183v2 Announce Type: replace Abstract: Vision-Language-Action (VLA) models have emerged as a promising approach for general-purpose robot manipulation. However, little research has mechan

model-releasesarxiv-cs-ro

9 Jun 2026

SpatialWorld: Benchmarking Interactive Spatial Reasoning of Multimodal Agents in Real-World Tasks

arXiv:2606.09669v1 Announce Type: new Abstract: Spatial reasoning is a foundational capability for multimodal large language models (MLLMs) to perceive and operate within the physical world. However,

SpectrumKV: Per-Token Mixed-Precision KV Cache Transfer for Prefill-Decode Disaggregated LLM Serving

arXiv:2606.08635v1 Announce Type: new Abstract: Prefill-decode (PD) disaggregation decouples prompt processing from token generation, but it also turns the key-value (KV) cache into a network payload.

SSAFE: Simple and Strong AI-Generated Image Detection via Frozen Vision Encoders

arXiv:2606.08634v1 Announce Type: new Abstract: The rapid advancement of generative models has blurred the boundary between synthetic and real imagery, creating an urgent need for reliable deepfake de

Stabilizing On-Policy Distillation for MLLM Reasoning with Global Normalization

arXiv:2606.09091v1 Announce Type: cross Abstract: On-policy distillation (OPD) has recently emerged as an important post-training paradigm. By using a stronger teacher model to provide dense, fine-gra

Still: Amortized KV Cache Compaction in a Single Forward Pass

arXiv:2606.07878v1 Announce Type: new Abstract: The KV cache is the memory bottleneck of long-horizon language model deployment. Practically, a deployable compactor must be lightweight enough to call

Storage Insights datasets: Enabling org-wide operational discovery with activity insights

model-releasesgoogle-cloud-ai

As enterprise storage footprints scale to billions of objects, AI applications and agentic workloads are fundamentally shifting the role of storage from a passive repository to the foundation of the d

Strained Coherence: A Pre-Failure Signal in Coding Agent Execution Trajectories

arXiv:2606.07889v1 Announce Type: cross Abstract: LLM-based coding agents sometimes acknowledge a problem in their own reasoning and then proceed anyway. We call this pattern strained coherence: a saf

Streaming Interventions: Can Video Large Language Models Correct Mistakes as They Occur?

arXiv:2606.09547v1 Announce Type: new Abstract: Learning everyday skills, like cooking a dish, relies increasingly on instructional media such as online videos. This opens the door to the use of video

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

Struct-Searcher: Agentic Structural Thinking Advances Multimodal Deep Information Seeking

arXiv:2606.07689v1 Announce Type: new Abstract: Deep research agents have attracted increasing attention for their ability to collect large-scale online information to acquire target knowledge, with r

Structured Neuron Pruning in Deep Neural Networks Using Multi-Armed Bandits

arXiv:2606.07615v1 Announce Type: cross Abstract: Deep neural networks often contain redundant hidden units. Removing individual weights can reduce parameter count, but unstructured sparsity is not al

Students without access to LLMs are 2 to 8 times more creative than students with access. That is the finding of a new paper comparing 2,200…

model-releasesgary-marcus--x

Students without access to LLMs are 2 to 8 times more creative than students with access. That is the finding of a new paper comparing 2,200 college admissions essays written by humans before ChatGPT

Subtitle-Aligned Fine-Tuning of Whisper for Swiss German ASR: Benchmark Contamination, Convention Mismatch, and an Honest Baseline at 25.6% WER (13.8% cWER)

arXiv:2606.07608v1 Announce Type: cross Abstract: We present a systematic study of fine-tuning OpenAI's Whisper large-v3 for Swiss German ASR, using 1,367 hours of broadcast speech paired with Standar

Supracompetitive Pricing Under AI Monoculture

arXiv:2601.01279v3 Announce Type: replace-cross Abstract: When competing sellers delegate pricing to a shared AI model, such as a large language model, correlated recommendations combined with perform

SurfDesign: Effective Protein Design on Molecular Surfaces

arXiv:2606.07567v1 Announce Type: cross Abstract: Protein function is largely determined by molecular surface geometry and physicochemical complementarity, yet most protein design methods condition on

SWE-Marathon: Can Agents Autonomously Complete Ultra-Long-Horizon Software Work?

arXiv:2606.07682v1 Announce Type: cross Abstract: AI agents are increasingly expected to complete long-horizon workflows that require sustained progress over hours, millions of tokens, and complex env

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

TABVERSE: Benchmarking Cross-Format Table Understanding in LLMs and VLMs

arXiv:2606.09578v1 Announce Type: new Abstract: Large Language Models (LLMs) and Vision-Language Models (VLMs) are increasingly evaluated on table reasoning tasks, but the role of table representation

TAME: A Trustworthy Test-Time Evolution of Agent Memory with Systematic Benchmarking

arXiv:2602.03224v2 Announce Type: replace Abstract: Test-time evolution of agent memory represents a pivotal paradigm for advancing AGI, as it strengthens complex reasoning through experience accumula

TempoBench: Evaluating Temporal Causal Reasoning in Large Language Models

arXiv:2510.27544v2 Announce Type: replace Abstract: Temporal reasoning involves understanding how systems evolve over time through input-driven state transitions. A key aspect is temporal causal reaso

The AI Epistemic Deference Index: A Continuous Measure of Sycophancy

arXiv:2606.07897v1 Announce Type: new Abstract: Current AI models frequently exhibit epistemic sycophancy, endorsing claims to agree with a user. Existing evaluations typically measure this either by

The Injection Paradox: Brand-Level Suppression in Safety-Trained LLM Recommendations via RAG Context Injection

arXiv:2606.09204v1 Announce Type: new Abstract: We present a reproducible failure mode of safety training in RAG-based LLM recommendation -- the Injection Paradox -- in which prompt injections embedde

The Last Visible Pixel: Probing Fine-Scale Perception in Vision-Language Models

arXiv:2606.07861v1 Announce Type: cross Abstract: Recent vision-language models (VLMs) excel at multimodal understanding and reasoning, yet their fine-grained visual perception remains underexplored.

The Montparnasse Algorithm for RNA Design

arXiv:2606.07562v1 Announce Type: cross Abstract: RNA design consists of discovering a nucleotide sequence that optimizes predefined criteria, such as secondary structure. It is useful for synthetic b

The Sample Complexity of Parameter-Free Stochastic Convex Optimization

arXiv:2506.11336v2 Announce Type: replace Abstract: We study the sample complexity of stochastic convex optimization when problem parameters such as the distance to optimality and the Lipschitz consta

TheoremBench: Evaluating LLMs on Theorem Proving in Formal Mathematics

arXiv:2606.09450v1 Announce Type: new Abstract: LLMs have recently achieved strong results on formal proving benchmarks. However, existing evaluations remain heavily concentrated on competition-style

Theoretical Foundations of Continual Learning via Drift-Plus-Penalty

arXiv:2606.08452v1 Announce Type: new Abstract: In many real-world settings, data streams are nonstationary and arrive sequentially, requiring learning systems to adapt continuously without retraining

There has been a lot of hand wringing on the appropriate valuation of SpaceX. Some large institutions believe SpaceX can only be valued at h…

model-releaseselon-musk--x

There has been a lot of hand wringing on the appropriate valuation of SpaceX. Some large institutions believe SpaceX can only be valued at half what the market seems to be willing to pay for it. Other

They ruled Iryna’s killer is incompetent to stand trial. The same system ruled this man was plenty competent enough to be released back into…

model-releaseselon-musk--x

They ruled Iryna’s killer is incompetent to stand trial. The same system ruled this man was plenty competent enough to be released back into society dozens of times. It’s past time to remove these lef

Thinking-Based Non-Thinking: Solving the Reward Hacking Problem in Training Hybrid Reasoning Models via Reinforcement Learning

arXiv:2601.04805v2 Announce Type: replace Abstract: Large reasoning models (LRMs) have attracted much attention due to their exceptional performance. However, their performance mainly stems from think

This is a super exciting release - Claude Fable 5 is the same underlying model as Mythos but with added safeguards. The benchmarks are great…

model-releaseskarpathy--x

This is a super exciting release - Claude Fable 5 is the same underlying model as Mythos but with added safeguards. The benchmarks are great and it's SOTA on everything by a margin but I'll add that *

Tiger Data launches PostgreSQL extension designed for AI agents

model-releasessiliconangle

Tiger Data today introduced a managed PostgreSQL database service designed specifically for AI agents, saying conventional database architectures are poorly suited to a future in which software is inc

Today, we released Gemini 3.5 Live Translate, our latest audio model for live speech-to-speech translation. It supports over 70 languages an…

model-releasesgoogle-ai--x

Today, we released Gemini 3.5 Live Translate, our latest audio model for live speech-to-speech translation. It supports over 70 languages and starts translating as soon as you start talking, streaming

Token Sample Complexity of Attention

arXiv:2512.10656v3 Announce Type: replace Abstract: As context windows in large language models continue to expand, it is essential to characterize how attention behaves at extreme sequence lengths. W

Towards Long-Horizon Vessel Trajectory and Destination Forecasting with Reasoning Large Language Models

arXiv:2606.08633v1 Announce Type: new Abstract: Long-horizon maritime trajectory prediction is important for shipping management, logistics planning, and maritime risk analysis, yet month-level foreca

Towards Personalized Bangla Book Recommendation: A Large-Scale Heterogeneous Book Graph Dataset

arXiv:2602.12129v2 Announce Type: replace-cross Abstract: Personalized book recommendation in Bangla literature has been constrained by the lack of structured, large-scale, and publicly available data

TQA-Bench: Evaluating LLMs for Multi-Table Question Answering

arXiv:2411.19504v2 Announce Type: replace Abstract: The advance of large language models (LLMs) has unlocked great opportunities in complex multi-modal data management tasks, particularly in question

Training-Free Generalized Few-Shot Segmentation through Open-Vocabulary Semantic Arbitration

arXiv:2606.09474v1 Announce Type: new Abstract: Generalized Few-Shot Semantic Segmentation (GFSS) has traditionally been approached as a representation-learning problem, requiring task-specific adapta

Trajectory-Refined Distillation

arXiv:2606.08432v1 Announce Type: new Abstract: On-policy distillation (OPD) has become a central post-training tool for large language models (LLMs), providing dense per-token teacher supervision alo

TriHead-GAN: A Generative Adversarial Network with Triple-Head Discriminator for Carbon Emission Time Series Generation

arXiv:2606.07569v1 Announce Type: new Abstract: Accurate carbon emission monitoring is critical for climate policy and emerging regulatory mechanisms such as the EU Carbon Border Adjustment Mechanism,

TRL-Bench: Standardizing Cross-Paradigm Representation-Level Evaluation of Tabular Encoders

arXiv:2606.09323v1 Announce Type: new Abstract: Tabular encoders are usually evaluated inside task-specific end-to-end pipelines, so models from different training paradigms are difficult to compare d

TT-DAC-PS: Twin-Target Deterministic Actor-Critic with Policy Smoothing for Optimal Trade Execution

arXiv:2606.08379v1 Announce Type: new Abstract: This study addresses the optimal execution of large stock sell programs by introducing TT-DAC-PS (Twin-Target Deterministic Actor-Critic with Policy Smo

Understanding Benchmark Language Under Weakened Formal Semantics

arXiv:2509.17455v2 Announce Type: replace-cross Abstract: State-of-the-art NLP benchmarks require interpretation of natural language that specifies conditions, procedures, and exceptions, often relyin

Understanding the Parameter Space Geometry of Transformers Encoding Boolean Functions

arXiv:2606.08768v1 Announce Type: new Abstract: Transformers consistently fail to learn certain simple functions that are provably expressible with specific parameter settings. This gap between learna

Unification of Closed-Open Industrial Detection Scenarios: New Large-Scale Benchmarks,Challenges and Baselines

arXiv:2606.07953v1 Announce Type: new Abstract: Large-scale Visual-Language Models (LVLMs) have achieved remarkable success in natural visual tasks, yet their application to industrial defect detectio

UniQL: Towards Dialect-Universal Benchmarking for Text-to-SQL

arXiv:2606.08018v1 Announce Type: new Abstract: Existing text-to-SQL benchmarks are largely centered on SQLite, making it difficult to evaluate whether models can generalize across heterogeneous SQL d