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 101–150 of 1135 papers

TitleStatusHype
NavGPT: Explicit Reasoning in Vision-and-Language Navigation with Large Language ModelsCode2
Precise Zero-Shot Dense Retrieval without Relevance LabelsCode2
mFollowIR: a Multilingual Benchmark for Instruction Following in RetrievalCode2
MentaLLaMA: Interpretable Mental Health Analysis on Social Media with Large Language ModelsCode2
GSCo: Towards Generalizable AI in Medicine via Generalist-Specialist CollaborationCode2
Long-Context Language Modeling with Parallel Context EncodingCode2
LVLM-eHub: A Comprehensive Evaluation Benchmark for Large Vision-Language ModelsCode2
LM-Nav: Robotic Navigation with Large Pre-Trained Models of Language, Vision, and ActionCode2
MAPLM: A Real-World Large-Scale Vision-Language Benchmark for Map and Traffic Scene UnderstandingCode2
Meta SecAlign: A Secure Foundation LLM Against Prompt Injection AttacksCode2
MineDreamer: Learning to Follow Instructions via Chain-of-Imagination for Simulated-World ControlCode2
BayLing: Bridging Cross-lingual Alignment and Instruction Following through Interactive Translation for Large Language ModelsCode2
Direct Preference Optimization of Video Large Multimodal Models from Language Model RewardCode2
DeSTA2: Developing Instruction-Following Speech Language Model Without Speech Instruction-Tuning DataCode2
Autonomous Improvement of Instruction Following Skills via Foundation ModelsCode2
Benchmarking Complex Instruction-Following with Multiple Constraints CompositionCode2
Aligning Modalities in Vision Large Language Models via Preference Fine-tuningCode2
ExpertPrompting: Instructing Large Language Models to be Distinguished ExpertsCode2
CuMo: Scaling Multimodal LLM with Co-Upcycled Mixture-of-ExpertsCode2
OmniBench: Towards The Future of Universal Omni-Language ModelsCode2
A Critical Evaluation of AI Feedback for Aligning Large Language ModelsCode2
LLaVAR: Enhanced Visual Instruction Tuning for Text-Rich Image UnderstandingCode2
FLASK: Fine-grained Language Model Evaluation based on Alignment Skill SetsCode2
DistiLLM-2: A Contrastive Approach Boosts the Distillation of LLMsCode2
LLM-RG4: Flexible and Factual Radiology Report Generation across Diverse Input ContextsCode2
AutoDefense: Multi-Agent LLM Defense against Jailbreak AttacksCode2
LLaVA-MORE: A Comparative Study of LLMs and Visual Backbones for Enhanced Visual Instruction TuningCode2
Aurora:Activating Chinese chat capability for Mixtral-8x7B sparse Mixture-of-Experts through Instruction-TuningCode2
DeSTA2.5-Audio: Toward General-Purpose Large Audio Language Model with Self-Generated Cross-Modal AlignmentCode2
LLaSM: Large Language and Speech ModelCode2
LLaVA-Plus: Learning to Use Tools for Creating Multimodal AgentsCode2
LMDrive: Closed-Loop End-to-End Driving with Large Language ModelsCode2
Critique Fine-Tuning: Learning to Critique is More Effective than Learning to ImitateCode2
LITA: Language Instructed Temporal-Localization AssistantCode2
LHRS-Bot-Nova: Improved Multimodal Large Language Model for Remote Sensing Vision-Language InterpretationCode2
Asynchronous RLHF: Faster and More Efficient Off-Policy RL for Language ModelsCode2
Lion: Adversarial Distillation of Proprietary Large Language ModelsCode2
Language Models are Homer Simpson! Safety Re-Alignment of Fine-tuned Language Models through Task ArithmeticCode2
CorDA: Context-Oriented Decomposition Adaptation of Large Language Models for Task-Aware Parameter-Efficient Fine-tuningCode2
Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free LunchCode2
Learning to Decode Collaboratively with Multiple Language ModelsCode2
AIR-Bench: Benchmarking Large Audio-Language Models via Generative ComprehensionCode2
CrystalFormer-RL: Reinforcement Fine-Tuning for Materials DesignCode2
Conifer: Improving Complex Constrained Instruction-Following Ability of Large Language ModelsCode2
Large Language Model Instruction Following: A Survey of Progresses and ChallengesCode2
EditWorld: Simulating World Dynamics for Instruction-Following Image EditingCode2
Agentic Reward Modeling: Integrating Human Preferences with Verifiable Correctness Signals for Reliable Reward SystemsCode2
EmoLLMs: A Series of Emotional Large Language Models and Annotation Tools for Comprehensive Affective AnalysisCode2
InFoBench: Evaluating Instruction Following Ability in Large Language ModelsCode2
CoIN: A Benchmark of Continual Instruction tuNing for Multimodel Large Language ModelCode2
Show:102550
← PrevPage 3 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