AI Paper Atlas

Research map / Benchmarks / ARC-Easy

ARC-Easy leaderboard

36 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on ARC-Easy. Best accuracy: 94.3 (Dynamic CoT (transfer), ICML 2025).

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.

ARC-Easy · accuracy

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

#accuracy ↑MethodPaperTopic
194.3Dynamic CoT (transfer)
GPT-4o-mini and GPT-4o
Token Signature: Predicting Chain-of-Thought Gains with Token Decoding Feature in Large Language Models
ICML 2025 · Table: Cross-model transfer ef
Reasoning and chain of thought
291.78PEG
Qwen2.5-7B-Instruct
Incentivizing Truthful Language Models via Peer Elicitation Games
NeurIPS 2025 · Table 1
Instruction and fine-tuning
385.8SV-Ensemble (M=16)
LLaMA-2-7B
Quantifying the Uncertainty of Foundation Models with Singular Value Ensembles
ICML 2026 · Table 3
RNNs and normalization
485.44FreeLB-RoBERTa (MTL)
RoBERTa-large
FreeLB: Enhanced Adversarial Training for Natural Language Understanding
ICLR 2020 · Table 3
Attacks on text and graphs
585.1Laplace-LoRA (LA)
LLaMA2-7B
Bayesian Low-rank Adaptation for Large Language Models
ICLR 2024 · Table 1
Instruction and fine-tuning
684.5Swimba-14BSwimba: Switch Mamba Model Scales State Space Models
NeurIPS 2026 · Table 1
Mixture of experts
784.21PMC
Gemma-3-12b-it
Model Collapse Is Not a Bug but a Feature in Machine Unlearning for LLMs
ICLR 2026 · Appendix utility table
Instruction and fine-tuning
883.54DPO w. LPC
Mistral-7B
Latent Preference Coding: Aligning Large Language Models via Discrete Latent Codes
ICML 2025 · Table 1
Alignment and preferences
983Ours (OP-SFT + KΔ-FT + FGPO)
Qwen2.5-7B
Feeling of Knowing in Large Language Models
NeurIPS 2026 · Table 2
Retrieval and knowledge
1080.26Anyprefer
LLaMA2-7B
Anyprefer: An Agentic Framework for Preference Data Synthesis
ICLR 2025 · Supplementary Experiments, Per
Alignment and preferences
26 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.