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 61–70 of 1135 papers

TitleStatusHype
Meta-Chunking: Learning Text Segmentation and Semantic Completion via Logical PerceptionCode3
ASFT: Aligned Supervised Fine-Tuning through Absolute LikelihoodCode3
ComfyBench: Benchmarking LLM-based Agents in ComfyUI for Autonomously Designing Collaborative AI SystemsCode3
1.5-Pints Technical Report: Pretraining in Days, Not Months -- Your Language Model Thrives on Quality DataCode3
NavGPT-2: Unleashing Navigational Reasoning Capability for Large Vision-Language ModelsCode3
AudioBench: A Universal Benchmark for Audio Large Language ModelsCode3
Self-play with Execution Feedback: Improving Instruction-following Capabilities of Large Language ModelsCode3
Refusal in Language Models Is Mediated by a Single DirectionCode3
MiLoRA: Harnessing Minor Singular Components for Parameter-Efficient LLM FinetuningCode3
FlashFace: Human Image Personalization with High-fidelity Identity PreservationCode3
Show:102550
← PrevPage 7 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