Every processed story in chronological order, with the newest coverage first. Filter by tag, source, or score to drill in.
SQA demonstrates that collective, diversity-enforced validator quorums can reduce unsafe LLM agent approvals in cloud infrastructure from 18.5% to 0.3%, addressing a safety gap that classical consensus protocols leave entirely unhandled.
The architecture provides formal, provable correctness guarantees for LLM agent executions — a property the paper demonstrates on regulated domains like healthcare billing compliance and security vulnerability disclosure where auditability is critical.
ALMANAC provides the first dataset with action-level mental model annotations grounded in authentic human collaboration, offering a concrete benchmark for evaluating whether LLM agents can simulate the reasoning alignment that effective human collaboration requires.
OLW targets a gap that the A2A spec itself acknowledges — standardized discovery registries — offering a queryable, structured alternative to the hardcoded agent relationships that currently characterize multi-agent systems.
The work demonstrates that an autonomous LLM-driven agent can produce physically interpretable, generalizable control policies through a fully auditable discovery process — without the black-box weight optimization that typically makes deep reinforcement learning opaque in scientific contexts.
The workflow demonstrates a concrete, cost-aware approach to composing multiple frontier models by phase — using each model where it outperforms the other — rather than relying on a single model for the entire development pipeline.
The paper identifies that active agent control over memory storage and retrieval — rather than passive, pipeline-fixed stores — is the key driver of cross-scenario generality, a finding that directly informs how memory systems for deployed LLM agents should be designed.
Lean4Agent introduces formal verification — previously absent from most agent systems — as a mechanism for specifying, debugging, and improving LLM agent workflows, with measured performance gains on established benchmarks.
Asuka-Bench exposes a dimension of code-agent capability — iterative repair from vague, evolving requirements — that existing one-shot benchmarks do not measure, and its unsaturated results (top model at 52%) show it remains a meaningful challenge for current LLMs.
The session offers a ground-level view from a major database vendor on the real blockers — stack choice, regulations, and evals — slowing enterprise AI agent adoption, grounded in MongoDB's direct experience serving frontier labs, AI-native startups, and large enterprises.