AI Paper Atlas

Research map / Benchmarks / MuSiQue

MuSiQue leaderboard

71 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on MuSiQue. Best EM: 54.59 (CF-RAG, 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.

MuSiQue · EM

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

#EM ↑MethodPaperTopic
154.59CF-RAG
Llama-3-8B-Instruct
Counterfactual Reasoning for Retrieval-Augmented Generation
ICLR 2026 · Table 1
Retrieval and knowledge
253HierSearch
Qwen2.5-7B-Instruct
HierSearch: A Hierarchical Enterprise Deep Search Framework Integrating Local and Web Searches
AAAI 2026 · Table 1
LLM agents
352.8CoopRAG (Gemma2-27B)
Gemma2-27B
Cooperative Retrieval-Augmented Generation for Question Answering: Mutual Information Exchange and Ranking by Contrasting Layers
NeurIPS 2025 · Table: QA performance comparis
Retrieval and knowledge
448.07Causal Prompting
LLaMA3
Causal Prompting: Debiasing Large Language Model Prompting Based on Front-Door Adjustment
AAAI 2025 · Main results table
Language, safety and interpretability
546INSES + Router
GLM-4
Beyond Explicit Edges: Robust Reasoning over Noisy and Sparse Knowledge Graphs
ICML 2026 · Table 2
Knowledge and temporal graphs
639.7OPERA (MAPGRPO)
Qwen2.5-7B
OPERA: A Reinforcement Learning--Enhanced Orchestrated Planner-Executor Architecture for Reasoning-Oriented Multi-Hop Retrieval
AAAI 2026 · Table 1
Retrieval and knowledge
738.5G-reasoner
GPT-4o-mini
G-reasoner: Foundation Models for Unified Reasoning over Graph-structured Knowledge
ICLR 2026 · Table 2
Retrieval and knowledge
837.2CodePlan
Mistral-7B
CodePlan: Unlocking Reasoning Potential in Large Language Models by Scaling Code-form Planning
ICLR 2025 · Table 1
Agents, code and math
936.6IRCoT + GFM-RAGGFM-RAG: Graph Foundation Model for Retrieval Augmented Generation
NeurIPS 2025 · Question answering performance
Knowledge and temporal graphs
1035.17PRIME
LLaMA3.3 70B
PRIME: Planning and Retrieval-Integrated Memory for Enhanced Reasoning
AAAI 2026 · Table 2
LLM agents
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

MuSiQue · F1

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

#F1 ↑MethodPaperTopic
166.36InfoRE
GPT-4
Information Re-Organization Improves Reasoning in Large Language Models
NeurIPS 2024 · Table 2
Retrieval and knowledge
264.23Causal Prompting
LLaMA3
Causal Prompting: Debiasing Large Language Model Prompting Based on Front-Door Adjustment
AAAI 2025 · Main results table
Language, safety and interpretability
358OPERA (MAPGRPO)
Qwen2.5-7B
OPERA: A Reinforcement Learning--Enhanced Orchestrated Planner-Executor Architecture for Reasoning-Oriented Multi-Hop Retrieval
AAAI 2026 · Table 2
Retrieval and knowledge
453.08SiReRAG (GPT-4o)SiReRAG: Indexing Similar and Related Information for Multihop Reasoning
ICLR 2025 · QA performance table
Retrieval and knowledge
552.5G-reasoner
GPT-4o-mini
G-reasoner: Foundation Models for Unified Reasoning over Graph-structured Knowledge
ICLR 2026 · Table 2
Retrieval and knowledge
652.3PANINI
GPT-4o-mini answer model
Panini: Continual Learning in Token Space via Structured Memory
ICML 2026 · Table 2
Retrieval and knowledge
748.9Psi-RAG (summative abstract)
Llama-3.3-70B
Hierarchical Abstract Tree for Cross-Document Retrieval Augmented Generation
ICML 2026 · Table 2
Human and embodied modeling
848.81PRIME
LLaMA3.3 70B
PRIME: Planning and Retrieval-Integrated Memory for Enhanced Reasoning
AAAI 2026 · Table 2
LLM agents
948.6HippoRAG 2
Llama-3.3-70B-Instruct
From RAG to Memory: Non-Parametric Continual Learning for Large Language Models
ICML 2025 · QA performance table
Retrieval and knowledge
1048.2XBridge
Llama-3.1-8B -> Qwen2.5-7B
XBRIDGE: Entity-Grounded Latent Bridge for Heterogeneous LLM Communication
NeurIPS 2026 · Table 1
LLM agents
25 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

MuSiQue · accuracy

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

#accuracy ↑MethodPaperTopic
174DCI-Agent-CC
Claude Sonnet 4.6
Beyond Semantic Similarity: Rethinking Retrieval for Agentic Search via Direct Corpus Interaction
NeurIPS 2026 · Table 1
Retrieval and knowledge
262.6KAR3
GPT-4 (1106-Preview)
From Complex to Atomic: Enhancing Augmented Generation via Knowledge-Aware Dual Rewriting and Reasoning
ICML 2025 · Table 1
Retrieval and knowledge
354.2C-3PO
Qwen2-72B
C-3PO: Compact Plug-and-Play Proxy Optimization to Achieve Human-like Retrieval-Augmented Generation
ICML 2025 · Main results table
LLM agents
440.8KVLink5
Llama-3.1-8B
KVLink: Accelerating Large Language Models via Efficient KV Cache Reuse
NeurIPS 2025 · Main results table
Reasoning and chain of thought
540Phi-MPO
Qwen3.5-9B
Agentic Multi-Turn Reasoning: A Fairness Approach
NeurIPS 2026 · Table 1
Reasoning and chain of thought
635.2WebSeer (Web Search)
Qwen2.5-14B
WebSeer: Training Deeper Search Agents through Reinforcement Learning with Self-Reflection
ICLR 2026 · Table 1
Human-AI and goal-conditioned
735.1HybridDeepSearcher
Qwen3-8B
Hybrid Deep Searcher: Scalable Parallel and Sequential Search Reasoning
ICLR 2026 · Table 2
Retrieval and knowledge
833.2Fathom-Search-4B (Stage-2)
Qwen3-4B
Fathom-DeepResearch: Unlocking Long Horizon Information Retrieval and Synthesis for SLMs
ICLR 2026 · Table 1
LLM agents
931.57Verifier-8x7B + Drafter-7B
Mistral-7B drafter; Mixtral-8x7B verifier
Speculative RAG: Enhancing Retrieval Augmented Generation through Drafting
ICLR 2025 · Main results table
Retrieval and knowledge
1026.02CDKC (w/ 2 round)
Llama-3.1-8B-Instruct
Know More, Know Clearer: A Meta-Cognitive Framework for Knowledge Augmentation in Large Language Models
ICML 2026 · Table 1
Retrieval and knowledge
4 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.