SOTAVerified

Instruction Following

Instruction following is the basic task of the model. This task is dedicated to evaluating the ability of the large model to follow human instructions. It is hoped that the model can generate controllable and safe answers.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 201–250 of 1135 papers

TitleStatusHype
NavGPT: Explicit Reasoning in Vision-and-Language Navigation with Large Language ModelsCode2
PandaGPT: One Model To Instruction-Follow Them AllCode2
ExpertPrompting: Instructing Large Language Models to be Distinguished ExpertsCode2
Lion: Adversarial Distillation of Proprietary Large Language ModelsCode2
Large Language Model Instruction Following: A Survey of Progresses and ChallengesCode2
Precise Zero-Shot Dense Retrieval without Relevance LabelsCode2
LM-Nav: Robotic Navigation with Large Pre-Trained Models of Language, Vision, and ActionCode2
The Replica Dataset: A Digital Replica of Indoor SpacesCode2
Habitat: A Platform for Embodied AI ResearchCode2
InstructTTSEval: Benchmarking Complex Natural-Language Instruction Following in Text-to-Speech SystemsCode1
Adversarial Paraphrasing: A Universal Attack for Humanizing AI-Generated TextCode1
RewardAnything: Generalizable Principle-Following Reward ModelsCode1
Incentivizing Reasoning for Advanced Instruction-Following of Large Language ModelsCode1
Don't Reinvent the Wheel: Efficient Instruction-Following Text Embedding based on Guided Space TransformationCode1
STRICT: Stress Test of Rendering Images Containing TextCode1
Speech-IFEval: Evaluating Instruction-Following and Quantifying Catastrophic Forgetting in Speech-Aware Language ModelsCode1
OmniGenBench: A Benchmark for Omnipotent Multimodal Generation across 50+ TasksCode1
IDA-Bench: Evaluating LLMs on Interactive Guided Data AnalysisCode1
AGENTIF: Benchmarking Instruction Following of Large Language Models in Agentic ScenariosCode1
Scaling Reasoning, Losing Control: Evaluating Instruction Following in Large Reasoning ModelsCode1
GIE-Bench: Towards Grounded Evaluation for Text-Guided Image EditingCode1
BLEUBERI: BLEU is a surprisingly effective reward for instruction followingCode1
MergeBench: A Benchmark for Merging Domain-Specialized LLMsCode1
A Multi-Dimensional Constraint Framework for Evaluating and Improving Instruction Following in Large Language ModelsCode1
MM-Skin: Enhancing Dermatology Vision-Language Model with an Image-Text Dataset Derived from TextbooksCode1
Adaptive Markup Language Generation for Contextually-Grounded Visual Document UnderstandingCode1
Instruction-Tuning Data Synthesis from Scratch via Web ReconstructionCode1
A Dual-Space Framework for General Knowledge Distillation of Large Language ModelsCode1
RealWebAssist: A Benchmark for Long-Horizon Web Assistance with Real-World UsersCode1
Sculpting Subspaces: Constrained Full Fine-Tuning in LLMs for Continual LearningCode1
STING-BEE: Towards Vision-Language Model for Real-World X-ray Baggage Security InspectionCode1
InsViE-1M: Effective Instruction-based Video Editing with Elaborate Dataset ConstructionCode1
Can Language Models Follow Multiple Turns of Entangled Instructions?Code1
REF-VLM: Triplet-Based Referring Paradigm for Unified Visual DecodingCode1
FuseChat-3.0: Preference Optimization Meets Heterogeneous Model FusionCode1
CrowdSelect: Synthetic Instruction Data Selection with Multi-LLM WisdomCode1
CodeIF: Benchmarking the Instruction-Following Capabilities of Large Language Models for Code GenerationCode1
StructFlowBench: A Structured Flow Benchmark for Multi-turn Instruction FollowingCode1
Enhancing Cross-Tokenizer Knowledge Distillation with Contextual Dynamical MappingCode1
IHEval: Evaluating Language Models on Following the Instruction HierarchyCode1
BenchMAX: A Comprehensive Multilingual Evaluation Suite for Large Language ModelsCode1
M-IFEval: Multilingual Instruction-Following EvaluationCode1
UltraIF: Advancing Instruction Following from the WildCode1
Online Preference Alignment for Language Models via Count-based ExplorationCode1
Curiosity-Driven Reinforcement Learning from Human FeedbackCode1
A Multi-Modal AI Copilot for Single-Cell Analysis with Instruction FollowingCode1
Facial Dynamics in Video: Instruction Tuning for Improved Facial Expression Perception and Contextual AwarenessCode1
Demystifying Domain-adaptive Post-training for Financial LLMsCode1
TinyHelen's First Curriculum: Training and Evaluating Tiny Language Models in a Simpler Language EnvironmentCode1
SPaR: Self-Play with Tree-Search Refinement to Improve Instruction-Following in Large Language ModelsCode1
Show:102550
← PrevPage 5 of 23Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1AutoIF (Llama3 70B)Inst-level loose-accuracy90.4—Unverified
2AutoIF (Qwen2 72B)Inst-level loose-accuracy88—Unverified
3GPT-4Inst-level loose-accuracy85.37—Unverified
4PaLM 2 SInst-level loose-accuracy59.11—Unverified