Where Does the Money Go Across Long-Running Coding Agents?

Researchers introduce WOVEN, a benchmark and training method that improves multimodal LLMs' visual transition reasoning capabilities by up to 27.3 percentage points on external benchmarks.

Researchers redesign 4-bit AdamW optimizer-state quantization with ZIP-SR and ZE-EDEN methods, reducing validation-loss gap to 32-bit AdamW by up to 70% in GPT- and Llama-style models.

Researchers developed white-box probes that detect LLM deception with up to 99.7% AUC, outperforming text-monitoring baselines.

Analyzes 2026 agent security incidents from OpenAI, Anthropic, and Google to propose a proactive security framework with boundary assurance stack.

VioLA trains humanoid control policies using human motion data by predicting motion latents that work for both humans and robots, achieving 100% zero-shot locomotion success on real hardware.