Professor, Computer Science Department at Stanford University.
Joined May 2011
- Tweets117
- Following42
- Followers2.5K
- Likes10
Monica Lam retweeted
Scientific 🔬 progress often starts with an unexpected observation that sparks a new hypothesis.
But AI agents often start with a hypothesis, then look for observations to support it.
We believe agents should actively explore both observations and hypotheses. When observations are only a byproduct of hypothesis testing, we risk missing discoveries we weren’t looking for.
⚡ Meet SHELDON: a scientific reasoning harness that models separate hypothesis and observation spaces and generalizes across medical diagnosis, system-log investigations, ARC-AGI 2, and even autoresearch.
Bazinga!
Monica Lam retweeted
OVAL has 2 papers at #COLM2026 presented tmr (Wed), both at Imperial Ballroom:
1. SatIR (SMT and Clinical Trial Matching) from 11:00 AM - 1:00PM, led by @cyruszzhou
2. DataSTORM (Deep Research + Database) at Imperial Ballroom #27 from 4:30 PM - 6:30 PM, led by @ShichengGLiu
Monica Lam retweeted
🏆We won the inaugural Databricks Grounded Reasoning Cup 🏆
Our team of @ShichengGLiu, @harshitj__, @PriyankShethia, and @JiudingSun took the first place, with the competition coming down to the final minute!
We're incredibly proud of the team, and the work behind the result!
A much-needed technique—match complex profiles to detailed requirements. High recall. Interpretable.
🔎 In high-stakes domains, a missed search result is not just a search error - it can mean a missed opportunity, delayed decision, or denied service.
Yet most retrieval systems still rank by keywords and embeddings. They can surface candidates that look relevant while missing those that actually satisfy the constraints.
What if an agent could turn those constraints into a retrieval tool that searches for viable options directly?
🌐 Project: satir.genie.stanford.edu/
📄 Paper: arxiv.org/abs/2604.08849
With @Yufei_1001, @kelakexyl, @YuChiangWang1, @ChiehJuChao1, and @MonicaSLam. Grateful to our collaborators across Stanford and Mayo Clinic, and to the broader @stanfordnlp and @StanfordHAI communities that helped shape the environment for this work.
#InformationRetrieval #AIAgents
<🧵1/n | 𝗦𝗮𝘁𝗜𝗥>
Monica Lam retweeted
📄 Paper: arxiv.org/abs/2509.23233
💻 Code & dataset: github.com/stanford-oval/inc…
Monica Lam retweeted
Takeaways:
- Contradictions are measurable and fixable at scale.
- LLMs aren't ready to fully automate yet (best AUROC 75.1% on WikiCollide) but are effective copilots.
Monica Lam retweeted
We're also releasing WikiCollide, a benchmark of real (not synthetic) Wikipedia inconsistencies.
Monica Lam retweeted
In a user study with experienced Wikipedia editors, CLAIRE led to 64.7% more inconsistencies found in the same time; 87.5% reported higher confidence.
Monica Lam retweeted
This Corpus-Level Inconsistency Detection task is needle-in-a-haystack hard.
Meet CLAIRE: an agent that surfaces potential contradictions with evidence and explanations.
Monica Lam retweeted
Excited to share our EMNLP 2025 (Main) paper: "Detecting Corpus-Level Knowledge Inconsistencies in Wikipedia with LLMs." How consistent is English Wikipedia? With the help of LLMs, we estimate 80M+ internally inconsistent facts (~3.3%). Small in percentage, large at corpus scale.
Monica Lam retweeted
The @StanfordOVAL is on a mission to make knowledge accessible to all. Its new open-source public AI assistant aims to empower researchers, journalists, and historians. See how it works IRL on Feb. 14: bit.ly/3WFb5K3
Monica Lam retweeted
Democratizing AI-Assisted Access to Knowledge!
The Stanford OVAL Lab is leading an initiative to create a public AI Assistant that democratizes access to the world's knowledge. Our pilot program, already embraced by over 400,000 users, generates Wikipedia-like articles through intelligent internet research.
In collaboration with historians and journalists, we are partnering with organizations such as @Wikimedia, @internetarchive, @CommonCrawl, and Data Commons to harness AI for accelerated research. (1/n)
Monica Lam retweeted
🌱Excited to introduce SPINACH, a Knowledge Base Question Answering agent & dataset on Wikidata, presented at EMNLP 2024! It combines LLMs, semantic parsing and graph traversal to set a new SOTA & is actively used by the Wikidata community.
ALT SPINACH: SPARQL-Based Information Navigation for Challenging Real-World Questions Shicheng Liu, Sina J. Semnani, Harold Triedman, Jialiang Xu, Isaac Dan Zhao, Monica S. Lam
Our Storm research project is reviewed in the Wikipedia research newsletter. Check it out.
In the latest issue of our newsletter:
A roundup of recent research on auto-creating Wikipedia-like articles using LLMs or Wikidata-based natural language generation, including STORM (AI agents role-play Wikipedia editors and subject experts) meta.wikimedia.org/wiki/Rese…
Monica Lam retweeted
Trending repository of the day 📈
storm by @StanfordOVAL/
An LLM-powered knowledge curation system that researches a topic and generates a full-length report with citations.
Last 24h: 958 ⭐
Total: 1900 ⭐️
github.com/stanford-oval/sto…
Monica Lam retweeted
🚨 New preprint 🚨
Current task-oriented AI assistants often struggle with reliability, control, and integrating knowledge. Even OpenAI's customer service bot uses a dialog tree-based assistant. We set out to solve this!
ALT LLM-Based Open-Domain Integrated Task and Knowledge Assistants with Programmable Policies from Harshit Joshi, Shicheng Liu, James Chen, Robert Weigle, Monica S. Lam.
Our WikiChat paper received the @wikimedia Research Award of the Year 2024! Try WikiChat at oval.cs.stanford.edu/, based on English, German, French, Spanish, Japanese, Russian, Portuguese, Italian, Chinese, Persian Wikipedias. Thanks to our sponsors @SloanFoundation & @Azure.
The @Wikimedia Research Award of the Year 2024 goes to "WikiChat: Stopping the hallucination of large language model chatbots by few-shot grounding on Wikipedia" ⚡
📜 aclanthology.org/2023.findin…
Monica Lam retweeted
My first in-person conference at #NAACL2024 🇲🇽
@_Yucheng_Jiang and I will present STORM🌪️ 𝘁𝗼𝗱𝗮𝘆 𝗮𝘁 𝟮𝗣𝗠 (📍Don Diego 2,3,4).
Would love to chat about report generation, lessons learned in hosting the web demo, and share a bit of what we are cooking after STORM!
Monica Lam retweeted
3 OVAL projects are awarded 2024-2025 Magic Grants!
“African History from the Bottom Up with LLM-Augmented Agents”, @sina_semnani et al.
“Cross-Lingual Multi-Perspective News”, @liamjxu et al.
“DataTalk: All Documents and Data, All at Once, All Verified”, @ShichengGLiu et al.
The happiest day of our year! Introducing the @BrownInstitute's 2024-2025 cohort of Magic Grant winners!
