Every processed story in chronological order, with the newest coverage first. Filter by tag, source, or score to drill in.
FrontierCode directly addresses a documented flaw in existing coding benchmarks — that passing tests does not equal mergeable code — by introducing maintainability-focused evaluation criteria that reveal current frontier models are far from solving real-world code quality.
Watch for the open-source release of SearchSwarm's harness, model weights, and training data, which could provide a practical foundation for building multi-agent deep research systems that scale beyond single-context-window limits.
Benchmark results showing even GPT-5 topping out at 17.4% TSR highlight how far current MLLMs are from reliable spatial reasoning, giving practitioners a rigorous testbed to measure progress on active exploration and long-horizon planning.
MCP server authors now have a concrete, public quality benchmark with actionable grade thresholds — and a badge system — to improve discoverability with agents.
Benchmark scores for coding agents are increasingly untrustworthy — CapCode and CapReward offer a concrete methodology for building evaluations and training regimes that resist shortcut exploitation and produce more honest capability measurements.
Benchmark your agentic tooling against these metrics — 87% time reduction and 55% lower dissatisfaction — as the paper establishes a concrete empirical baseline for what autonomous end-to-end execution delivers over conversational search in real production settings.
Audit every step of a complex AI research pipeline — the explicit traceability and rubric-grounded synthesis in DuMate-DeepResearch offer a concrete blueprint for reducing hallucination and improving accountability in agentic coding and research systems.
Fine-tuning on DragOn's 3.5M drag-grounding tasks offers a concrete path to improving GUI agent accuracy on complex interactions — like resizing, highlighting, and slider control — that current models handle poorly.
Treat eval score gains as a diagnostic signal rather than a leaderboard goal — Khan's three-zone failure-analysis framework gives AI/coding practitioners a concrete method for extracting actionable improvements from broken benchmarks without overfitting to them.
Sparse attention research bottlenecks slow both human researchers and AI coding agents — Vortex's programmable serving layer removes that friction, enabling faster automated exploration of attention algorithms for long-context LLM deployments.