Every processed story in chronological order, with the newest coverage first. Filter by tag, source, or score to drill in.
MiniPIC removes the requirement for identical prefixes to reuse KV cache entries, enabling efficient caching of recurring structured inputs in retrieval-augmented and agentic workloads without the large server-side code changes or host-to-device transfer overhead of prior PIC approaches.
Connai replaces the per-project rebuild of context retrieval and OAuth integrations with a single shared vector DB, letting agents reason across application boundaries through one MCP endpoint rather than stitching together independent per-app configs.
The integration demonstrates a concrete pattern where scoping MCP access to read-only unlocks natural-language business analysis against live operational data without requiring users to navigate a dashboard.
The server makes a 150-year corpus of international soccer data instantly queryable by any MCP-compatible AI agent without credentials or infrastructure setup, demonstrating a zero-friction pattern for shipping domain-specific RAG corpora as MCP servers.
The shared-daemon architecture eliminates the per-client ~400 MB embedding model load, meaning multiple Claude windows share a single in-memory model instance rather than each paying the full RAM cost independently.
The paper addresses a core limitation of existing LLM agent memory systems — difficulty with evidence aggregation and fact revision across sessions — by introducing a structured, maintainable architecture that improves both how memory is organized and how it is retrieved.
LakeQA exposes a significant performance gap in frontier LLMs — including GPT-5.2 at 18.37% exact-match — on tasks that require jointly searching a massive heterogeneous data lake and performing multi-hop reasoning, a combination absent from prior comprehensive benchmarks.
The server addresses two concrete pain points for AI research agents — hitting Semantic Scholar's strict rate limits and exhausting context windows — by combining a discovery-first retrieval strategy with local caching and resilient concurrency controls.
OSF represents a concrete implementation of micropayment-gated, citation-backed data access for AI agents, directly addressing the verifiability gap that arises when agents rely on scraped or RAG-retrieved content from unattributed sources.
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.