AI Paper Atlas

Research map / Benchmarks / WebShop

WebShop leaderboard

39 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on WebShop. Best success rate: 99 (W2SG with MCTS, ICLR 2026).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

WebShop · success rate

39 papers, each with its best reported number for its own method; higher is better.

#success rate ↑MethodPaperTopic
199W2SG with MCTS
Llama-2-13B strong, Llama-2-7B weak
Weak-to-Strong Generalization with Failure Trajectories
ICLR 2026 · Table 1
Visual and representation RL
288.9DAPO+AEM
Qwen2.5-7B-Instruct
AEM: Adaptive Entropy Modulation for Multi-Turn Agentic Reinforcement Learning
NeurIPS 2026 · Table 1
RL from feedback
386.5RLOO w/ iStar
Qwen2.5-7B-Instruct
Agentic Reinforcement Learning with Implicit Step Rewards
ICLR 2026 · Table 1
RL from feedback
485.7Dr.BoT (GiGPO) + SPEAR
Qwen2.5-7B-Instruct
Learn the Ropes, Then Trust the Wins: Self-imitation with Progressive Exploration for Agentic Reinforcement Learning
ICLR 2026 · Table 1
Visual and representation RL
585.2SCOPE
Llama3.1-8B
From Local Skills to Long-Horizon Tasks: Progressive Skill Exploration for LLM Web Agents
NeurIPS 2026 · Table 3
LLM agents
684.2StraTA
Qwen2.5-7B-Instruct
StraTA: Incentivizing Agentic Reinforcement Learning with Strategic Trajectory Abstraction
NeurIPS 2026 · Table 1
Visual and representation RL
783.3HiPER
Qwen2.5-7B-Instruct
HiPER: Hierarchical Plan–Execute RL for Multi-Turn LLM Agents
ICML 2026 · Table 1
LLM agents
882.8GiGPO + PA-MoE
Qwen2.5-7B-Instruct
Phase-Aware Mixture of Experts for Agentic Reinforcement Learning
ICML 2026 · Table 1
Visual and representation RL
982.7DAPO + EMPG
Qwen2.5-7B-Instruct
Harnessing Uncertainty: Entropy-Modulated Policy Gradients for Long-Horizon LLM Agents
ICML 2026 · Table 1
RL from feedback
1082.42T2PO
Qwen3-8B-RFT
T$^2$PO: Uncertainty-Guided Exploration Control for Stable Multi-Turn Agentic Reinforcement Learning
ICML 2026 · Table 1
RL from feedback
29 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

WebShop · score

13 papers, each with its best reported number for its own method; higher is better.

3 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.