AI Paper Atlas

Research map / Benchmarks / DROP

DROP leaderboard

14 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on DROP. Best accuracy: 91.9 (Targeted alternation training, ICML 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.

DROP · accuracy

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

#accuracy ↑MethodPaperTopic
191.9Targeted alternation training
GPT-4.1-mini
Learning Rewrite-Invariant Reasoning with Targeted Alternation Training
ICML 2026 · Table 1
Instruction and fine-tuning
289.62InfiGFusion
Phi-4
InfiGFusion: Graph-on-Logits Distillation via Efficient Gromov-Wasserstein for Model Fusion
NeurIPS 2025 · Main results table
Reasoning and chain of thought
385.3PASR
Qwen3-8B
A Stitch in Time Saves Nine: Proactive Self-Refinement for Language Models
ICLR 2026 · Table 1
Language, safety and interpretability
482.55VeriTime
Qwen2.5-3B-Instruct
Time Series Reasoning via Process-Verifiable Thinking Data Synthesis and Scheduling for Tailored LLM Reasoning
ICML 2026 · Table 5
Foundation models for time series
582.28MASPOB
GPT-4o-mini
MASPOB: Bandit-Based Prompt Optimization for Multi-Agent Systems with Graph Neural Networks
ICML 2026 · Table 1
LLM agents
680LEAP high-level
GPT-4-turbo
In-Context Principle Learning from Mistakes
ICML 2024 · Textual Reasoning results tabl
Language, safety and interpretability
765.42OptimSyn (Ours)
Qwen3-8B-Base
OptimSyn: Influence-Guided Rubrics Optimization for Synthetic Data Generation
ICLR 2026 · Table 1
Instruction and fine-tuning
840.5SoT
Llama-3.1-8B
State of Thought Enables Endogenous Reasoning
NeurIPS 2026 · Table 1
Reasoning and chain of thought
935.7ICLMIn-Context Pretraining: Language Modeling Beyond Document Boundaries
ICLR 2024 · Reading comprehension table
Language, safety and interpretability
1035AugCon
Llama3-70B-Instruct
Automatically Generating Numerous Context-Driven SFT Data for LLMs Across Diverse Granularity
AAAI 2025 · Automatic evaluation table
Instruction and fine-tuning
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.