Andy Liu
@andyliu.bsky.social
phd type things @ cmu lti
andyjliu.github.io
andyjliu.github.io
Pinned
Andy Liu
@andyliu.bsky.social
· Oct 2
🚨New Paper: LLM developers aim to align models with values like helpfulness or harmlessness. But when these conflict, which values do models choose to support? We introduce ConflictScope, a fully-automated evaluation pipeline that reveals how models rank values under conflict.
(📷 xkcd)
(📷 xkcd)
Reposted by Andy Liu
🚨New paper: Reward Models (RMs) are used to align LLMs, but can they be steered toward user-specific value/style preferences?
With EVALUESTEER, we find even the best RMs we tested exhibit their own value/style biases, and are unable to align with a user >25% of the time. 🧵
With EVALUESTEER, we find even the best RMs we tested exhibit their own value/style biases, and are unable to align with a user >25% of the time. 🧵
October 14, 2025 at 3:59 PM
🚨New paper: Reward Models (RMs) are used to align LLMs, but can they be steered toward user-specific value/style preferences?
With EVALUESTEER, we find even the best RMs we tested exhibit their own value/style biases, and are unable to align with a user >25% of the time. 🧵
With EVALUESTEER, we find even the best RMs we tested exhibit their own value/style biases, and are unable to align with a user >25% of the time. 🧵
🚨New Paper: LLM developers aim to align models with values like helpfulness or harmlessness. But when these conflict, which values do models choose to support? We introduce ConflictScope, a fully-automated evaluation pipeline that reveals how models rank values under conflict.
(📷 xkcd)
(📷 xkcd)
October 2, 2025 at 4:04 PM
🚨New Paper: LLM developers aim to align models with values like helpfulness or harmlessness. But when these conflict, which values do models choose to support? We introduce ConflictScope, a fully-automated evaluation pipeline that reveals how models rank values under conflict.
(📷 xkcd)
(📷 xkcd)
Placing LLMs in simulated markets helps us quantitatively and qualitatively measure their propensity to collude, as well as how environmental changes affect this. Read below or find @veronateo.bsky.social at the ICML multi-agent systems workshop to learn more!
Excited to share our paper “Evaluating LLM Agent Collusion in Double Auctions”!
We put LLMs in a simulated market and find that collusion increases when they are able to communicate via natural language, differs across models, and is influenced by urgency and oversight.
1/
We put LLMs in a simulated market and find that collusion increases when they are able to communicate via natural language, differs across models, and is influenced by urgency and oversight.
1/
July 9, 2025 at 1:24 PM
Placing LLMs in simulated markets helps us quantitatively and qualitatively measure their propensity to collude, as well as how environmental changes affect this. Read below or find @veronateo.bsky.social at the ICML multi-agent systems workshop to learn more!
Reposted by Andy Liu
CMU LTI is hosting predoc interns this summer, centered around "Language Technologies for All"! Please apply and circulate! lti.cs.cmu.edu/news-and-eve...
CMU LTI Language Technology for All Internship 2025 - Language Technologies Institute - School of Computer Science - Carnegie Mellon University
The LTI is currently seeking applicants for the summer 2025 Language Technology for All Internship
lti.cs.cmu.edu
January 7, 2025 at 10:42 PM
CMU LTI is hosting predoc interns this summer, centered around "Language Technologies for All"! Please apply and circulate! lti.cs.cmu.edu/news-and-eve...
looking for 2025 book recs!
things i've previously liked, for reference -
nonfiction: the structure of scientific revolutions, cybernetic revolutionaries, seeing like a state
fiction: stories of your life and others, one hundred years of solitude, project hail mary, recursion
things i've previously liked, for reference -
nonfiction: the structure of scientific revolutions, cybernetic revolutionaries, seeing like a state
fiction: stories of your life and others, one hundred years of solitude, project hail mary, recursion
January 3, 2025 at 9:58 PM
looking for 2025 book recs!
things i've previously liked, for reference -
nonfiction: the structure of scientific revolutions, cybernetic revolutionaries, seeing like a state
fiction: stories of your life and others, one hundred years of solitude, project hail mary, recursion
things i've previously liked, for reference -
nonfiction: the structure of scientific revolutions, cybernetic revolutionaries, seeing like a state
fiction: stories of your life and others, one hundred years of solitude, project hail mary, recursion
Reposted by Andy Liu
Looking for all your LTI friends on Bluesky? The LTI Starter Pack is here to help!
go.bsky.app/NhTwCVb
go.bsky.app/NhTwCVb
November 20, 2024 at 4:15 PM
Looking for all your LTI friends on Bluesky? The LTI Starter Pack is here to help!
go.bsky.app/NhTwCVb
go.bsky.app/NhTwCVb