AI Paper Atlas

Research map / Benchmarks / ARC-Challenge

ARC-Challenge leaderboard

53 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on ARC-Challenge. Best accuracy: 96.33 (FLARE-9B (AR-Trust), NeurIPS 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.

ARC-Challenge · accuracy

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

#accuracy ↑MethodPaperTopic
196.33FLARE-9B (AR-Trust)
Qwen3.5-9B
FLARE: Diffusion for Hybrid Language Model
NeurIPS 2026 · Table 1
Language, safety and interpretability
294.5SpecExit
Qwen3-4B-Thinking-2507
SpecExit: Accelerating Large Reasoning Model via Speculative Exit
ICML 2026 · Table 1
Reasoning and chain of thought
394AgentCDM cross-dataset (trained on MMLU_PRO)
Qwen-7B-Base
AgentCDM: Enhancing Multi-Agent Collaborative Decision-Making via ACH-Inspired Structured Reasoning
AAAI 2026 · Cross-dataset table
LLM agents
491.7Skill-MoESkill-Based Mixture-of-Experts: Adaptive Routing for Heterogeneous Reasoning via Inferred Skills
ICML 2026 · Appendix ARC table
Agents, code and math
590.8ALC(9,2)
Qwen2.5-7B (LoRA)
On the Construction and Implications of Low-Loss Valleys in LoRA-based Bayesian Inference
NeurIPS 2026 · Table 1
Pruning and efficient tuning
688.7Self-consistency
PaLM-540B
Self-Consistency Improves Chain of Thought Reasoning in Language Models
ICLR 2023 · Table 2
Reasoning and chain of thought
788.2LazyStack-SubUnifying Stacking and Cascading for Efficient Ensemble Inference
ICML 2026 · text
RNNs and normalization
887.01PEG
Qwen2.5-7B-Instruct
Incentivizing Truthful Language Models via Peer Elicitation Games
NeurIPS 2025 · Table 1
Instruction and fine-tuning
986.6AlphaRouter (2.1B active)
Qwen2.5-0.5B + Qwen2.5-7B
AlphaRouter: Token-level Routing Between SLM and LLM with Reinforcement Learning and Tree Search
ICML 2026 · Table 1
RL from feedback
1085Markovian
Qwen3-4B
Markovian Transformers for Informative Language Modeling
ICLR 2026 · Table 6
Reasoning and chain of thought
43 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.