AI Paper Atlas

Research map / Large language models / Agents, code and math

Training LLM Agents to Empower Humans

Evan Ellis, Vivek Myers, Jens Tuyls, Sergey Levine, Anca Dragan, Benjamin Eysenbach

ICML 2026 · Large language models › Agents, code and math · approach: Agent instruction-following benchmarks · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
train with Empower logit-threshold completions instead of SFT-20Llama3.1-8B Instruct assistant · LiveCodeBench (Llama-3.3-70B simulated human)Pass@1: 0.097 -> 0.282Improved
train with Empower instead of untrained BaseLlama3.1-8B Instruct assistant · LiveCodeBench (Llama-3.3-70B simulated human)Pass@1: 0.170 -> 0.282Improved
train with Empower instead of untrained BaseQwen3-14B assistant · LiveCodeBench (Gemma simulated human)Pass@1: 0.086 -> 0.170Improved
train with Empower instead of untrained BaseQwen3-8B assistant · LiveCodeBench (Gemma simulated human)Pass@1: 0.092 -> 0.178Improved
cap suggestion length at 10 tokens (Base-10) instead of Base (ablation)Qwen3-8B assistant · LiveCodeBench (Llama-3.3-70B simulated human)Accept Ratio: 0.351 -> 0.592Improved
cap suggestion length at 10 tokens (Base-10) instead of Base (ablation)Qwen3-8B assistant · LiveCodeBench (Llama-3.3-70B simulated human)Pass@1: 0.183 -> 0.198Mixed
cap suggestion length at 10 tokens (Base-10) instead of Base (ablation)Llama3.1-8B Instruct assistant · LiveCodeBench (Gemma simulated human)Pass@1: 0.064 -> 0.064No clear effect
cap suggestion length at 10 tokens (Base-10) instead of Base (ablation)Qwen3-14B assistant · LiveCodeBench (Gemma simulated human)Pass@1: 0.086 -> 0.062Hurt
supervised finetune on next 20 human tokens (SFT-20) instead of BaseLlama3.1-8B Instruct assistant · LiveCodeBench (Llama-3.3-70B simulated human)Pass@1: 0.170 -> 0.097Hurt
use Empower assistant instead of Base-20 with real usersLlama-3.1-8B assistant · Human studyacceptance rate: 6.18% -> 8.08%Improved
use Empower assistant instead of Base-20 with real usersLlama-3.1-8B assistant · Human studycharacters deleted per accepted suggestion: 12.91 -> 9.56Improved
use Empower assistant instead of Base-20 with real usersLlama-3.1-8B assistant · Human studymost relevant suggestions preference: 61% (p = 0.240)No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy