Picture for Jacob Andreas

Jacob Andreas

ThinkPrune: Pruning Long Chain-of-Thought of LLMs via Reinforcement Learning

Add code
Apr 02, 2025
Viaarxiv icon

(How) Do Language Models Track State?

Add code
Mar 04, 2025
Viaarxiv icon

LM Agents for Coordinating Multi-User Information Gathering

Add code
Feb 17, 2025
Viaarxiv icon

The Surprising Effectiveness of Test-Time Training for Abstract Reasoning

Add code
Nov 11, 2024
Figure 1 for The Surprising Effectiveness of Test-Time Training for Abstract Reasoning
Figure 2 for The Surprising Effectiveness of Test-Time Training for Abstract Reasoning
Figure 3 for The Surprising Effectiveness of Test-Time Training for Abstract Reasoning
Figure 4 for The Surprising Effectiveness of Test-Time Training for Abstract Reasoning
Viaarxiv icon

LoRA vs Full Fine-tuning: An Illusion of Equivalence

Add code
Oct 28, 2024
Figure 1 for LoRA vs Full Fine-tuning: An Illusion of Equivalence
Figure 2 for LoRA vs Full Fine-tuning: An Illusion of Equivalence
Figure 3 for LoRA vs Full Fine-tuning: An Illusion of Equivalence
Figure 4 for LoRA vs Full Fine-tuning: An Illusion of Equivalence
Viaarxiv icon

A Hitchhiker's Guide to Scaling Law Estimation

Add code
Oct 15, 2024
Viaarxiv icon

MisinfoEval: Generative AI in the Era of "Alternative Facts"

Add code
Oct 15, 2024
Figure 1 for MisinfoEval: Generative AI in the Era of "Alternative Facts"
Figure 2 for MisinfoEval: Generative AI in the Era of "Alternative Facts"
Figure 3 for MisinfoEval: Generative AI in the Era of "Alternative Facts"
Figure 4 for MisinfoEval: Generative AI in the Era of "Alternative Facts"
Viaarxiv icon

Learning Linear Attention in Polynomial Time

Add code
Oct 14, 2024
Viaarxiv icon

Learning How Hard to Think: Input-Adaptive Allocation of LM Computation

Add code
Oct 07, 2024
Viaarxiv icon

Algorithmic Capabilities of Random Transformers

Add code
Oct 06, 2024
Viaarxiv icon