Professor, MIT EECS and Chief Scientist, Cambridge Mobile Telematics
- Tweets 358
- Following 435
- Followers 3,414
- Likes 440
I've been thinking about the impact of AI on education and spent some time trying out the latest AI detection tools this weekend.
linkedin.com/pulse/pitfalls-…
1
5
481
Cool new work from my student Joshua.
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.
🚨 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 "𝘱𝘦𝘦𝘬𝘴 𝘪𝘯𝘵𝘰" these environments.
🚀 Our paper, 𝗣𝗘𝗘𝗞: 𝙖 𝙨𝙮𝙨𝙩𝙚𝙢 𝙛𝙤𝙧 𝙗𝙪𝙞𝙡𝙙𝙞𝙣𝙜 𝙖𝙣𝙙 𝙢𝙖𝙞𝙣𝙩𝙖𝙞𝙣𝙞𝙣𝙜 𝗮𝗻 𝗼𝗿𝗶𝗲𝗻𝘁𝗮𝘁𝗶𝗼𝗻 𝗰𝗮𝗰𝗵𝗲 𝙛𝙤𝙧 𝙇𝙇𝙈 𝙖𝙜𝙚𝙣𝙩𝙨, introduces this idea.
Compared with strong baselines, including RAG, Compaction Agents, and SOTA prompt-learning frameworks, PEEK dominates the cost–quality Pareto frontier: achieving +6.3–34.0% 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)
4
23
7,033
Sam Madden retweeted
📢 CfP: Conference on Innovative Data Systems Research #CIDR2027 (Amsterdam, The Netherlands). Papers due: August 4, 2026. @ViktorLeis, @samrmadden, and I are looking forward to your paper & demo submissions! cidrdb.org/cidr2027/
3
14
946
Interesting article about #SemBench, a new benchmark for semantic AI operators; our Palimpsest system does super well on it!
📣 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!
📜Paper: bit.ly/3WGsZf6
💻Website: sembench.org
💾Code: bit.ly/49DeuAg
1
789
Start thinking about your CIDR submissions now!
📢 CfP: Conference on Innovative Data Systems Research #CIDR2026 (Chaminade, USA). Papers due: August 5, 2025. Gustavo Alonso, @samrmadden, and I are looking forward to your paper & demo submissions! cidrdb.org/cidr2026/
1
2
1,408
Sam Madden retweeted
Henry Corrigan-Gibbs, a professor in @MITEECS, has been named a recipient of the 2024 Junior Bose Award.
5
11
1,305
Last chance to register for CIDR 2025 in Amsterdam!
The CIDR 2025 program is online: cidrdb.org/cidr2025/program.…
Registration closes today -- Join us in Amsterdam:
cidrdb.org/cidr2025/registra…
1
4
1,827
Sam Madden retweeted
Vol:17 No:12 → Databases Unbound: Querying All of the World’s Bytes with AI vldb.org/pvldb/vol17/p4546-m…
1
5
30
4,103
Sam Madden retweeted
It took three years to finish, but our follow-up to the 2006 "What Goes Around Comes Around" is finally out! Stonebraker and I examine the last 20 years in databases and discuss why relational databases + SQL will continue to remain on top.
📄PDF: db.cs.cmu.edu/papers/2024/wh…
ALT What Goes Around Comes Around… And Around… SIGMOD Record 2024
23
333
1,254
173,146
Sam Madden retweeted
🚨 NEW EPISODE: High Impact in Databases with... Joe Hellerstein (@joe_hellerstein) is available now!
🎧 Listen on Apple ➡️ podcasts.apple.com/us/podcas…
🎧 Listen on Spotify ➡️ open.spotify.com/show/6IQIF9…
☕️ Support the podcast ➡️ buymeacoffee.com/disseminate
1
3
17
5,826
Excited to announce the release of our Palimpzest system and paper!
If this (very high-level) summary of our work has piqued your interest -- go read our full paper!
📄Paper: arxiv.org/pdf/2405.14696
💻Code: github.com/mitdbg/palimpzest…
We would love to hear any feedback, ideas for more use cases, and/or opportunities for collaboration.
9
1,572
Sam Madden retweeted
Submit to #NEDB2024 by 4/12 (11:59pm EST): cmt3.research.microsoft.com/…
* Talks (2-page abstracts) on recent/new research or industry experience
* Posters (1-page abstract)
Stay tuned for more details on registration!
bu-disc.github.io/nedbday/20…
@samrmadden @vkalavri @__ssarkar
4
6
2,401
Sam Madden retweeted
Amazon Redshift Serverless is launching next gen AI-driven scaling and optimizations. With the great work of @tim_kraska and his team, Redshift pushes the capabilities of serverless data analytics!
#amazon #redshift #serverless #aidrivenscaling
6
16
3,412
Sam Madden retweeted
It's official — our company, @ponderdata, is all set to join Snowflake! So delighted that we're joining forces with the leading cloud data warehouse — and bringing our data science capabilities to all their customers.
Snowflake announced its intent to acquire @ponderdata to further enable Python data scientists in the Data Cloud. We look forward to welcoming the Ponder team and the Modin community to Snowflake. Learn more: okt.to/I6WXlc
26
15
220
47,058
Sam Madden retweeted
What are the most influential database papers of all time? Everyone has an opinion, but here's some data!
I ran PageRank on the VLDB/SIGMOD/CIDR citation graph. Search for a paper, view a specific year, or look up an author's most "influential" papers.
rmarcus.info/blog/2023/07/25…
7
23
115
17,112
Sam Madden retweeted
Paper is available here: github.com/tli2/tli2.github.…
DARQ is joint work with @badrishc @sebburckhardt @samrmadden and will soon be available in open source as part of the Microsoft FASTER project.
Come hang out on Wednesday for the talk if you are interested!
1
7
1,704
By the way, registration deadline is this Wednesday! Hope to see you all there!
Program is up and registration is now open for Northeast Database Day, March 10, 2023, at Northeastern University in Boston!
northeastern-datalab.github.…
1
1
5
2,276