🔴 High Significance
Model Releases
🔴 Large Language Models Explore by Latent Distilling — score 75
Sources: huggingface
Generating diverse responses is crucial for test-time scaling of large language models (LLMs), yet standard stochastic sampling mostly yields surface-level lexical variation, limiting semantic exploration. In this paper, we propose Exploratory Sampling (ESamp), a decoding approach that explicitly en
Developer Tools
🔴 GLM-5V-Turbo: Toward a Native Foundation Model for Multimodal Agents — score 95
Sources: huggingface
We present GLM-5V-Turbo, a step toward native foundation models for multimodal agents. As foundation models are increasingly deployed in real environments, agentic capability depends not only on language reasoning, but also on the ability to perceive, interpret, and act over heterogeneous contexts s
🔴 RADIO-ViPE: Online Tightly Coupled Multi-Modal Fusion for Open-Vocabulary Semantic SLAM in Dynamic Environments — score 85
Sources: huggingface
We present RADIO-ViPE (Reduce All Domains Into One -- Video Pose Engine), an online semantic SLAM system that enables geometry-aware open-vocabulary grounding, associating arbitrary natural language queries with localized 3D regions and objects in dynamic environments. Unlike existing approaches tha
🟡 Notable
Model Releases
🟡 ClawGym: A Scalable Framework for Building Effective Claw Agents — score 65
Sources: huggingface
Claw-style environments support multi-step workflows over local files, tools, and persistent workspace states. However, scalable development around these environments remains constrained by the absence of a systematic framework, especially one for synthesizing verifiable training data and integratin
🟡 Introducing Advanced Account Security — score 50
Sources: lab_blog/OpenAI
Introducing Advanced Account Security: phishing-resistant login, stronger recovery, and enhanced protections to safeguard sensitive data and prevent account takeover.
🟡 Enabling a new model for healthcare with AI co-clinician — score 50
Sources: lab_blog/DeepMind
Researching the path to AI-augmented care and development of an AI co-clinician.
Developer Tools
🟡 Turning the TIDE: Cross-Architecture Distillation for Diffusion Large Language Models — score 55
Sources: huggingface
Diffusion large language models (dLLMs) offer parallel decoding and bidirectional context, but state-of-the-art dLLMs require billions of parameters for competitive performance. While existing distillation methods for dLLMs reduce inference steps within a single architecture, none address cross-arch
🟡 Diffusion Templates: A Unified Plugin Framework for Controllable Diffusion — score 45
Sources: huggingface
Controllable diffusion methods have substantially expanded the practical utility of diffusion models, but they are typically developed as isolated, backbone-specific systems with incompatible training pipelines, parameter formats, and runtime hooks. This fragmentation makes it difficult to reuse inf
🟢 Incremental
Model Releases
🟢 Operating-Layer Controls for Onchain Language-Model Agents Under Real Capital — score 10
Sources: huggingface
We study reliability in autonomous language-model agents that translate user mandates into validated tool actions under real capital. The setting is DX Terminal Pro, a 21-day deployment in which 3,505 user-funded agents traded real ETH in a bounded onchain market. Users configured vaults through str
Developer Tools
🟢 FAMA: Failure-Aware Meta-Agentic Framework for Open-Source LLMs in Interactive Tool Use Environments — score 35
Sources: huggingface
Large Language Models are being increasingly deployed as the decision-making core of autonomous agents capable of effecting change in external environments. Yet, in conversational benchmarks, which simulate real-world customer-centric issue resolution scenarios, these agents frequently fail due to t
🟢 Accelerating RL Post-Training Rollouts via System-Integrated Speculative Decoding — score 25
Sources: huggingface
RL post-training of frontier language models is increasingly bottlenecked by autoregressive rollout generation, making rollout acceleration a central systems challenge. Many existing efficiency methods improve throughput by changing the rollout or optimization regime, for example, through off-policy
🟢 Unified 4D World Action Modeling from Video Priors with Asynchronous Denoising — score 10
Sources: huggingface
We propose X-WAM, a Unified 4D World Model that unifies real-time robotic action execution and high-fidelity 4D world synthesis (video + 3D reconstruction) in a single framework, addressing the critical limitations of prior unified world models (e.g., UWM) that only model 2D pixel-space and fail to
📄 New Papers
| Title | Category | Score | Link |
|---|---|---|---|
| GLM-5V-Turbo: Toward a Native Foundation Model for Multimodal Agents | developer_tool | 89 | Open |
| RADIO-ViPE: Online Tightly Coupled Multi-Modal Fusion for Open-Vocabulary Semantic SLAM in Dynamic Environments | developer_tool | 67 | Open |
| Large Language Models Explore by Latent Distilling | model_release | 65 | Open |
| ClawGym: A Scalable Framework for Building Effective Claw Agents | model_release | 48 | Open |
| Turning the TIDE: Cross-Architecture Distillation for Diffusion Large Language Models | developer_tool | 44 | Open |
| The Inverse-Wisdom Law: Architectural Tribalism and the Consensus Paradox in Agentic Swarms | cs.AI | 0 | Open |
| Evaluating Epistemic Guardrails in AI Reading Assistants: A Behavioral Audit of a Minimal Prototype | cs.AI | 0 | Open |
| BrainDINO: A Brain MRI Foundation Model for Generalizable Clinical Representation Learning | cs.AI | 0 | Open |
| Learning When to Remember: Risk-Sensitive Contextual Bandits for Abstention-Aware Memory Retrieval in LLM-Based Coding Agents | cs.AI | 0 | Open |
| Mechanized Foundations of Structural Governance: Machine-Checked Proofs for Governed Intelligence | cs.AI | 0 | Open |
| The Two Boundaries: Why Behavioral AI Governance Fails Structurally | cs.AI | 0 | Open |
| Learning Rate Engineering: From Coarse Single Parameter to Layered Evolution | cs.AI | 0 | Open |
| Machine Collective Intelligence for Explainable Scientific Discovery | cs.AI | 0 | Open |
| METASYMBO: Multi-Agent Language-Guided Metamaterial Discovery via Symbolic Latent Evolution | cs.AI | 0 | Open |
| BoostLoRA: Growing Effective Rank by Boosting Adapters | cs.AI | 0 | Open |