A Chinese AI company could land its model on major US cloud platforms for the first time, taking a cut of the revenue. The article Chinese Moonshot AI…
ResiSpec: Enhancing Multi-Candidate Speculative Sampling via Residual Distribution Shaping
arXiv:2608.24411v1 Announce Type: new Abstract: The efficiency of Large Language Model (LLM) serving is fundamentally limited by the sequential nature of…
Can You Defend What Your AI Just Did?
Regulatory Deadlines for AI Keep Slipping, but the Need for AI Accountability Remains For two years, the regulatory conversation around enterprise AI has…
Do Recipes Have Personas? Characterizing and Generating Creator Style in Attributed Procedural Graphs
arXiv:2608.24369v1 Announce Type: new Abstract: While large language models (LLMs) possess vast zero-shot procedural knowledge, their tendency to produce…
AI News Brief Hourly Summary 2026-08-26 14h : 16 posts
16 posts published in the last hour 11:33Can a Dynamic Internal Field Govern a Transformer’s Cognition? Certifiability, not Superiority, in Homeostatic Compute Control 11:33Benchmarking LLM Judges for Voice-Agent Evaluation: Reliability, Calibration, and Human Oversight 11:33Runable hits $21M to bet AI…
Can a Dynamic Internal Field Govern a Transformer’s Cognition? Certifiability, not Superiority, in Homeostatic Compute Control
arXiv:2608.24319v1 Announce Type: new Abstract: An intelligent system does not merely reason: it governs its own reasoning – how much to compute, when to…
Benchmarking LLM Judges for Voice-Agent Evaluation: Reliability, Calibration, and Human Oversight
arXiv:2608.24314v1 Announce Type: new Abstract: Evaluating conversational voice agents at scale re- quires reliable assessment methods that capture both…
Runable hits $21M to bet AI agents can go from building businesses to growing them
Runable says 60%–70% of its 1 trillion-plus token usage in the last 90 days came from paying customers.
SonarLLM: A Native Sonar–Optical Multimodal Large Language Model for Underwater Perception
arXiv:2608.24325v1 Announce Type: new Abstract: Reliable underwater perception requires complementary sensing under variable visibility. Optical cameras…
Insilico Medicine Posts First Profit, $106.3M Revenue in First Half of 2026
The clinical-stage generative AI drug discovery company recorded a net profit of $35.54 million and an adjusted net profit of $51.23 million, with…
Selective Regenerative Decoding: Trajectory-Level Intervention for Inference-Time Reasoning
arXiv:2608.24338v1 Announce Type: new Abstract: Inference-time decoding methods improve LLM reasoning by exploring multiple candidate trajectories, yet…
SCX.ai Partners With DDN to Scale Australia’s Sovereign AI Inference Cloud
Australian sovereign AI infrastructure company SCX.ai and data intelligence vendor DDN announced a partnership on August 24, 2026, to expand what the…
The Handoff Tax: Continuing Non-Native Trajectories in LLM Agents
arXiv:2608.24358v1 Announce Type: new Abstract: Coding agents perform long-running tasks spanning dozens of model calls, tool uses, and code edits. As…
ReproAgent: Contract-Guided Paper-to-Code Reproduction
arXiv:2608.24291v1 Announce Type: new Abstract: Paper-to-code reproduction asks scientific AI agents to turn research papers into executable repositories…
Eating for a Sustainable Planet: Personalized Sustainable Diet Recommendation via Constraint-Aware Decision-Making Modeling
arXiv:2608.24274v1 Announce Type: new Abstract: A sustainable diet represents a multi-dimensional synergy among four essential pillars: nutrition…
RePolicy: Reinforcement Learning for Safety-Policy Invocation in Agent Safeguards
arXiv:2608.24275v1 Announce Type: new Abstract: Safeguarding language model agents requires assessing complete execution trajectories under…
IBM drops open-weight Granite 4.2 family with built-in agentic capabilities under Apache 2.0
IBM is releasing its Granite 4.2 language models in 3B, 8B, and 30B sizes, trained on about 15 trillion tokens with a context window of up to 512,000…
VideoHarness-RSI: Recursive Harness Self-Improvement for Long-Video Understanding with Frozen Vision-Language Models
arXiv:2608.24302v1 Announce Type: new Abstract: Long-video understanding depends critically on how a limited model context is constructed from a much…