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 1111–1120 of 1135 papers

TitleStatusHype
EIFBENCH: Extremely Complex Instruction Following Benchmark for Large Language ModelsCode0
MM-R5: MultiModal Reasoning-Enhanced ReRanker via Reinforcement Learning for Document RetrievalCode0
AbsInstruct: Eliciting Abstraction Ability from LLMs through Explanation Tuning with Plausibility EstimationCode0
Does Alignment Tuning Really Break LLMs' Internal Confidence?Code0
Continual Learning for Instruction Following from Realtime FeedbackCode0
Phased Instruction Fine-Tuning for Large Language ModelsCode0
Generative Visual Instruction TuningCode0
Semantic Graphs for Syntactic Simplification: A Revisit from the Age of LLMCode0
Stay Focused: Problem Drift in Multi-Agent DebateCode0
Style Outweighs Substance: Failure Modes of LLM Judges in Alignment BenchmarkingCode0
Show:102550
← PrevPage 112 of 114Next →

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