AI Paper Atlas

Research map / Benchmarks / BBH

BBH leaderboard

49 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on BBH. Best accuracy: 94 (AgentNet, NeurIPS 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.

BBH · accuracy

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

#accuracy ↑MethodPaperTopic
194AgentNet
DeepSeek-V3
AgentNet: Decentralized Evolutionary Coordination for LLM-based Multi-Agent Systems
NeurIPS 2025 · Main Results table
LLM agents
293.4HiVA
Qwen2.5-72B-Instruct-Turbo
HiVA: Self-organized Hierarchical Variable Agent via Goal-driven Semantic-Topological Evolution
AAAI 2026 · Table 1
LLM agents
390TypedThinker + SC @5
GPT-4o
TypedThinker: Diversify Large Language Model Reasoning with Typed Thinking
ICLR 2025 · Table: GPT-4o
Agents, code and math
485.8SynLogic-32B
Qwen2.5-32B-Base
SynLogic: Synthesizing Verifiable Reasoning Data at Scale for Learning Logical Reasoning and Beyond
NeurIPS 2025 · Table 1
Agents, code and math
585.62InfiGFusion
Phi-4
InfiGFusion: Graph-on-Logits Distillation via Efficient Gromov-Wasserstein for Model Fusion
NeurIPS 2025 · Main results table
Reasoning and chain of thought
684.2DOTS: External
GPT-4o-mini
DOTS: Learning to Reason Dynamically in LLMs via Optimal Reasoning Trajectories Search
ICLR 2025 · External Planner Tuning Result
Reasoning and chain of thought
781GPT-4 + Self-Discover
GPT-4
SELF-DISCOVER: Large Language Models Self-Compose Reasoning Structures
NeurIPS 2024 · Main results table
Reasoning and chain of thought
878FLV-RL
Qwen2.5-14B
Pushing the Boundaries of Natural Reasoning: Interleaved Bonus from Formal-Logic Verification
ICML 2026 · Table 1
Agents, code and math
976.6GReaTer
Gemma-2-9B
GReaTer: Gradients Over Reasoning Makes Smaller Language Models Strong Prompt Optimizers
ICLR 2025 · Table 1
Language, safety and interpretability
1075.65OptimSyn (Ours)
Qwen3-8B-Base
OptimSyn: Influence-Guided Rubrics Optimization for Synthetic Data Generation
ICLR 2026 · Table 1
Instruction and fine-tuning
27 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

BBH · score

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

#score ↑MethodPaperTopic
184.07Avengers
10 open-source ~7B models
The Avengers: A Routing Recipe for Collective Intelligence in Language Models
AAAI 2026 · Table 2
Language, safety and interpretability
268.9IF-RLVR Tulu-DPO 8B
Tulu-DPO
Generalizing Verifiable Instruction Following
NeurIPS 2025 · Table: Model performances on d
Agents, code and math
354.31StressPrompt level 6
Phi-3-mini-4k-Instruct
StressPrompt: Does Stress Impact Large Language Models and Human Performance Similarly?
AAAI 2025 · Table 1
AI and society
454.3NRT-GM
Llama-3.1-8B
Native Reasoning Models: Training Language Models to Reason on Unverifiable Data
ICLR 2026 · Table 3
Reasoning and chain of thought
552.8IAA-8 (1.2M)
Llama3-8B
IAA: Inner-Adaptor Architecture Empowers Frozen Large Language Model with Multimodal Capabilities
AAAI 2025 · Table: Comparison on Text-only
Multimodal LLMs
651.38MPO Iter.4
Llama-3-8B-SFT
Magnetic Preference Optimization: Achieving Last-iterate Convergence for Language Model Alignment
ICLR 2025 · Table: Evaluation results on O
Alignment and preferences
748.9SelectIT
LLaMA-2-13B
SelectIT: Selective Instruction Tuning for LLMs via Uncertainty-Aware Self-Reflection
NeurIPS 2024 · Table 1
Instruction and fine-tuning
848.78DIL w/ LSIF
Llama3-8B-Base
On a Connection Between Imitation Learning and RLHF
ICLR 2025 · Table 1
Alignment and preferences
944.6SlowFast Sampling
LLaDA 8B
Accelerating Diffusion Large Language Models with SlowFast Sampling: The Three Golden Principles
ICLR 2026 · Table 1
Language, safety and interpretability
1043.55Cal-DPO
Zephyr-7B-SFT
Cal-DPO: Calibrated Direct Preference Optimization for Language Model Alignment
NeurIPS 2024 · Table: UltraFeedback Binarized
Alignment and preferences
2 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.