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 51–100 of 1135 papers

TitleStatusHype
When Large Multimodal Models Confront Evolving Knowledge:Challenges and PathwaysCode2
Don't Reinvent the Wheel: Efficient Instruction-Following Text Embedding based on Guided Space TransformationCode1
Differential Information: An Information-Theoretic Perspective on Preference Optimization—0
ARC: Argument Representation and Coverage Analysis for Zero-Shot Long Document Summarization with Instruction Following LLMs—0
ChartMind: A Comprehensive Benchmark for Complex Real-world Multimodal Chart Question Answering—0
LaMDAgent: An Autonomous Framework for Post-Training Pipeline Optimization via LLM Agents—0
Adaptive Detoxification: Safeguarding General Capabilities of LLMs through Toxicity-Aware Knowledge Editing—0
Let Them Talk: Audio-Driven Multi-Person Conversational Video GenerationCode7
A Course Correction in Steerability Evaluation: Revealing Miscalibration and Side Effects in LLMsCode0
PartInstruct: Part-level Instruction Following for Fine-grained Robot Manipulation—0
Evaluating Robustness of Large Audio Language Models to Audio Injection: An Empirical Study—0
From Alignment to Advancement: Bootstrapping Audio-Language Alignment with Synthetic Data—0
Leveraging Importance Sampling to Detach Alignment Modules from Large Language Models—0
StyleAR: Customizing Multimodal Autoregressive Model for Style-Aligned Text-to-Image Generation—0
RECAST: Strengthening LLMs' Complex Instruction Following with Constraint-Verifiable Data—0
STRICT: Stress Test of Rendering Images Containing TextCode1
Speech-IFEval: Evaluating Instruction-Following and Quantifying Catastrophic Forgetting in Speech-Aware Language ModelsCode1
Optimal Transport-Based Token Weighting scheme for Enhanced Preference OptimizationCode0
OmniGenBench: A Benchmark for Omnipotent Multimodal Generation across 50+ TasksCode1
MIDB: Multilingual Instruction Data Booster for Enhancing Multilingual Instruction Synthesis—0
IDA-Bench: Evaluating LLMs on Interactive Guided Data AnalysisCode1
LIFEBench: Evaluating Length Instruction Following in Large Language ModelsCode0
IFEval-Audio: Benchmarking Instruction-Following Capability in Audio-based Large Language ModelsCode3
CASTILLO: Characterizing Response Length Distributions of Large Language ModelsCode0
LaViDa: A Large Diffusion Language Model for Multimodal UnderstandingCode3
ToDi: Token-wise Distillation via Fine-Grained Divergence Control—0
AGENTIF: Benchmarking Instruction Following of Large Language Models in Agentic ScenariosCode1
In-Context Watermarks for Large Language Models—0
ManipLVM-R1: Reinforcement Learning for Reasoning in Embodied Manipulation with Large Vision-Language Models—0
Sparse Activation Editing for Reliable Instruction Following in Narratives—0
Diffusion vs. Autoregressive Language Models: A Text Embedding Perspective—0
ThinkLess: A Training-Free Inference-Efficient Method for Reducing Reasoning Redundancy—0
FlowKV: Enhancing Multi-Turn Conversational Coherence in LLMs via Isolated Key-Value Cache Management—0
Hunyuan-TurboS: Advancing Large Language Models through Mamba-Transformer Synergy and Adaptive Chain-of-Thought—0
Joint Flashback Adaptation for Forgetting-Resistant Instruction Tuning—0
Scaling Reasoning, Losing Control: Evaluating Instruction Following in Large Reasoning ModelsCode1
DecIF: Improving Instruction-Following through Meta-Decomposition—0
Ground-V: Teaching VLMs to Ground Complex Instructions in Pixels—0
Two Experts Are All You Need for Steering Thinking: Reinforcing Cognitive Effort in MoE Reasoning Models Without Additional Training—0
Domain Adaptation of VLM for Soccer Video Understanding—0
Causal Head Gating: A Framework for Interpreting Roles of Attention Heads in Transformers—0
What Prompts Don't Say: Understanding and Managing Underspecification in LLM PromptsCode0
Rethinking Predictive Modeling for LLM Routing: When Simple kNN Beats Complex Learned Routers—0
Multi-Level Aware Preference Learning: Enhancing RLHF for Complex Multi-Instruction Tasks—0
KIT's Offline Speech Translation and Instruction Following Submission for IWSLT 2025—0
CompBench: Benchmarking Complex Instruction-guided Image Editing—0
Internal Causal Mechanisms Robustly Predict Language Model Out-of-Distribution BehaviorsCode0
Enhancing Complex Instruction Following for Large Language Models with Mixture-of-Contexts Fine-tuning—0
GIE-Bench: Towards Grounded Evaluation for Text-Guided Image EditingCode1
HelpSteer3-Preference: Open Human-Annotated Preference Data across Diverse Tasks and Languages—0
Show:102550
← PrevPage 2 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