Cédric Colas

Postdoc@Massachusetts Institute of Technology·美国·OpenReview

研究方向

open-endeness · language model · agent · social learning · cultural learning · exploration · intrinsic motivations · curiosity · reinforcement learning · goal-directed learning

6.4

Language and Experience: A Computational Model of Social Learning in Complex Tasks

ICLR 2026Poster

一作

3.5

Learning Progress-Guided LLM Goal Generation for Autotelic Skill Learning

ICLR 2026Rejected

3.0

Code-enabled language models can outperform reasoning models on diverse tasks

ICLR 2026Rejected

二作

7.3

Self-Improving Language Models for Evolutionary Program Synthesis: A Case Study on ARC-AGI

ICML 2025Poster

二作

6.0

When LLMs Play the Telephone Game: Cultural Attractors as Conceptual Tools to Evaluate LLMs in Multi-turn Settings

ICLR 2025Poster

4.9

MAGELLAN: Metacognitive predictions of learning progress guide autotelic LLM agents in large goal spaces

合作者 (20)