AI Paper Atlas

Research map / Benchmarks / Countdown

Countdown leaderboard

13 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on Countdown. Best accuracy: 90.5 (TMLT r=32, 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.

Countdown · accuracy

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

#accuracy ↑MethodPaperTopic
190.5TMLT r=32On Expressive Power of Looped Transformers: Theoretical Analysis and Enhancement via Timestep Encoding
ICML 2025 · Table 1
Attention and transformers
285.9SPRINT
DeepSeek-R1-Distill-Qwen-7B
SPRINT: Enabling Interleaved Planning and Parallelized Execution in Reasoning Models
NeurIPS 2025 · Results table
Reasoning and chain of thought
385RandOpt
OLMo3-7B-Instruct
Neural Thickets: Diverse Task Experts Are Dense Around Pretrained Weights
ICML 2026 · Table 4
Mixture of experts
478.9CoRP
OLMo3-7B-Instruct
Consolidating Rewarded Perturbations for LLM Post-Training
NeurIPS 2026 · Table 1
Reasoning and chain of thought
571.48UFT
Qwen2.5-1.5B
UFT: Unifying Supervised and Reinforcement Fine-Tuning
NeurIPS 2025 · Table: Performance comparison
Instruction and fine-tuning
666.8ES
Qwen-2.5-7B-Instruct
Evolution Strategies at Scale: LLM Fine-Tuning Beyond Reinforcement Learning
ICML 2026 · Table 1
Instruction and fine-tuning
763.1GRPO + coupled rhythm credit
Qwen3-4B-Base
Attention Illuminates LLM Reasoning: The Uncovered Preplan-and-Anchor Rhythm Enables Fine-Grained Policy Optimization
ICML 2026 · text
Reasoning and chain of thought
856.6d2-StepMerge
LLaDA-8B-Instruct
d2: Improving Reasoning in Diffusion Language Models via Trajectory Likelihood Estimation
ICML 2026 · Table 4
Reasoning and chain of thought
940.8PC-sampler + POKE-SMC (M=4)
LLaDA-8B-Instruct
Lookahead Path Likelihood Optimization for Diffusion LLMs
ICML 2026 · Table 1
Reasoning and chain of thought
1033.2LLaDA + WINO
LLaDA-8B-Instruct
Wide-In, Narrow-Out: Revokable Decoding for Efficient and Effective DLLMs
ICLR 2026 · Table 1
Language, safety and interpretability
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.