Grace Liu

29 posts

Grace Liu

Grace Liu

@GraceLiu78

Katılım Ağustos 2024
4 Takip Edilen150 Takipçiler
Grace Liu retweetledi
Michiel Bakker
Michiel Bakker@bakkermichiel·
🚨📄 New preprint! We find the “boiling the frog” equivalent of AI use. In a series of RCTs, we show that after just 10 min of AI assistance people perform worse and give up more often than those who never used AI. w Grace Liu @brianchristian Mira Dumbalska and Rachit Dubey 🧵
Michiel Bakker tweet media
English
27
242
737
137.8K
Grace Liu retweetledi
Mahsa Bastankhah
Mahsa Bastankhah@MBastankhah·
We will be presenting our poster “demystifying the mechanism behind emergent exploration in goal conditioned RL” today @iclr_conf Time: 3:15-5:45 pm Location: Pavilion 4 P4-#3404 @ben_eysenbach @GraceLiu78 @Dilip_Arumugam
Grace Liu@GraceLiu78

NEW PAPER: "Demystifying the Mechanisms Behind Emergent Exploration in Goal-conditioned RL" How do RL algorithms develop sophisticated exploration strategies without explicit rewards? We provide insight into this question by studying Single-Goal Contrastive RL (SGCRL). [1/9]

English
0
7
32
3.1K
Grace Liu retweetledi
Chongyi Zheng
Chongyi Zheng@chongyiz1·
1/ Reinforcement learning is usually framed as maximizing rewards. But can we cast it as reaching the right goals? New blog on bridging RL, goal-conditioned RL, and stochastic shortest path: iclr-blogposts.github.io/2026/blog/2026… Also #ICLR2026 Poster: Thu 10:30 AM–1:00 PM, P4 #4611. 🧵⬇️
Chongyi Zheng tweet media
English
2
26
146
22.6K
Grace Liu
Grace Liu@GraceLiu78·
Explore our paper for more insights: ⛏️ How counterfactual pairs teach models to recognize information sufficiency? ⛏️ Why verbal reasoning acts as an implicit value function for termination? ⛏️ How reasoning stabilizes decision boundaries and improves OOD robustness? [8/9]
English
1
0
2
498
Grace Liu
Grace Liu@GraceLiu78·
NEW PAPER: "CaRT: Teaching LLM Agents to Know When They Know Enough"! LLMs often overthink, ask too many questions, or waste compute. We introduce Counterfactuals and Reasoning for Termination (CaRT) - teaching LLMs when to stop gathering info and make decisions. 🧵[1/9]
Grace Liu tweet media
English
1
12
38
13.7K
Grace Liu
Grace Liu@GraceLiu78·
✅ Takeaway: Understanding WHY algorithms work helps us build safer, more reliable AI systems. Emergent exploration isn’t magic ✨ — it’s shaped representations doing exactly what the optimization objective prescribes. 🎯 [8/9]
English
1
1
6
872
Grace Liu
Grace Liu@GraceLiu78·
NEW PAPER: "Demystifying the Mechanisms Behind Emergent Exploration in Goal-conditioned RL" How do RL algorithms develop sophisticated exploration strategies without explicit rewards? We provide insight into this question by studying Single-Goal Contrastive RL (SGCRL). [1/9]
English
2
9
66
20.4K

Keşfet

@brianchristian @iclr_conf @ben_eysenbach @Dilip_Arumugam @princeton_rl @MBastankhah @QuYuxiao @aviral_kumar2