Twitxy Saved

Sam Madden

@samrmadden

Professor, MIT EECS and Chief Scientist, Cambridge Mobile Telematics
437 Following 3.4K Followers 360 Posts

Latest posts 5

Sam Madden @samrmadden
Cool new work from my student Joshua.
Joshua Gu @astrogu_ May 20
Recent agentic systems (Claude Code, Codex, RLM, etc.) push context out of the prompt and into the environment (e.g., as files). This helps them maintain long-term knowledge about their goals and functionality.

๐Ÿšจ While this is a good idea, we show a surprising result: systems that use external environments like this perform much better when given a small, fixed-size, in-context, agent-managed cache that "๐˜ฑ๐˜ฆ๐˜ฆ๐˜ฌ๐˜ด ๐˜ช๐˜ฏ๐˜ต๐˜ฐ" these environments.

๐Ÿš€ Our paper, ๐—ฃ๐—˜๐—˜๐—ž: ๐™– ๐™จ๐™ฎ๐™จ๐™ฉ๐™š๐™ข ๐™›๐™ค๐™ง ๐™—๐™ช๐™ž๐™ก๐™™๐™ž๐™ฃ๐™œ ๐™–๐™ฃ๐™™ ๐™ข๐™–๐™ž๐™ฃ๐™ฉ๐™–๐™ž๐™ฃ๐™ž๐™ฃ๐™œ ๐—ฎ๐—ป ๐—ผ๐—ฟ๐—ถ๐—ฒ๐—ป๐˜๐—ฎ๐˜๐—ถ๐—ผ๐—ป ๐—ฐ๐—ฎ๐—ฐ๐—ต๐—ฒ ๐™›๐™ค๐™ง ๐™‡๐™‡๐™ˆ ๐™–๐™œ๐™š๐™ฃ๐™ฉ๐™จ, introduces this idea.

Compared with strong baselines, including RAG, Compaction Agents, and SOTA prompt-learning frameworks, PEEK dominates the costโ€“quality Pareto frontier: achieving +6.3โ€“34.0% in quality, with fewer iterations and lower cost.

Paper: arxiv.org/abs/2605.19932
GitHub: github.com/zhuohangu/peek

More in the thread below! (1/N)
View quoted post on X
0 replies 4 reposts 22 likes 7.1K views View on X
Sam Madden @samrmadden
Interesting article about #SemBench, a new benchmark for semantic AI operators; our Palimpsest system does super well on it!
Matthew Russo Verified account @RussoMatthew Nov 6, 2025
๐Ÿ“ฃ We're spreading the word about #SemBench -- a brand new benchmark for semantic query processing over multimodal workloads including text, image, audio, and tabular data!

๐Ÿ“œPaper: bit.ly/3WGsZf6
๐Ÿ’ปWebsite: sembench.org
๐Ÿ’พCode: bit.ly/49DeuAg
View quoted post on X