Every processed story in chronological order, with the newest coverage first. Filter by tag, source, or score to drill in.
CSTS addresses a core bottleneck in agentic LLM development by replacing manual skill engineering with an automated, multi-model collective process that explicitly tests whether skills transfer across models — a property the paper identifies as critical for robust generalization.
mcp-gen removes the need to manually write MCP schemas by deriving them directly from TypeScript type definitions.
ACCORD demonstrates that a training-free grounding layer can close a substantial portion of the task-completion gap in LLM agents across both digital and embodied benchmarks, without modifying the underlying model.
The release fixes a silent data-loss bug in `Sandbox.getMetrics()` where time-range parameters were ignored, and closes a correctness gap where empty-body error responses were swallowed rather than surfaced.
The tutorial provides a concrete, reproducible starting point for the agentic post-training workflow — SFT from agent traces — before the more complex GRPO and environment RL stages that follow in the series.
Gemma 4's availability on Bedrock gives developers managed access to Apache 2.0-licensed open-weight models with native function calling and multimodal support across dense and MoE architectures.
The double iframe architecture is the direct result of ruling out every simpler sandboxing approach, meaning MCP app developers who understand the constraint can anticipate the strict domain-declaration requirement and avoid submission rejections.
Strands Evals provides structured, automated root cause analysis for AI agent failures — including confidence scores, causal chains, and targeted fix recommendations — replacing ad-hoc manual debugging in evaluation pipelines.
ASSAY demonstrates that matching skills to tasks at inference time — rather than global library curation — is the key bottleneck for experience-based agent improvement, achieving state-of-the-art results on two benchmarks without any weight updates.
DeepRoot is the first system to simultaneously achieve low hallucination rates (7–10%) and high reasoning coherence on historical medical text, demonstrating a viable path for converting pre-ontological archives into verifiable drug-discovery leads at scale.