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 226–250 of 1135 papers

TitleStatusHype
HalluciDoctor: Mitigating Hallucinatory Toxicity in Visual Instruction DataCode1
Defending Large Language Models Against Jailbreaking Attacks Through Goal PrioritizationCode1
DeCoRe: Decoding by Contrasting Retrieval Heads to Mitigate HallucinationsCode1
AlpaGasus: Training A Better Alpaca with Fewer DataCode1
Guiding Multi-Step Rearrangement Tasks with Natural Language InstructionsCode1
DANLI: Deliberative Agent for Following Natural Language InstructionsCode1
AlpaCare:Instruction-tuned Large Language Models for Medical ApplicationCode1
GIE-Bench: Towards Grounded Evaluation for Text-Guided Image EditingCode1
Demystifying Domain-adaptive Post-training for Financial LLMsCode1
LIONs: An Empirically Optimized Approach to Align Language ModelsCode1
Generation-driven Contrastive Self-training for Zero-shot Text Classification with Instruction-following LLMCode1
Curiosity-Driven Reinforcement Learning from Human FeedbackCode1
DialFRED: Dialogue-Enabled Agents for Embodied Instruction FollowingCode1
Generative Parameter-Efficient Fine-TuningCode1
AllenAct: A Framework for Embodied AI ResearchCode1
CrowdSelect: Synthetic Instruction Data Selection with Multi-LLM WisdomCode1
Bactrian-X: Multilingual Replicable Instruction-Following Models with Low-Rank AdaptationCode1
Lexicon Learning for Few-Shot Neural Sequence ModelingCode1
Cross-model Control: Improving Multiple Large Language Models in One-time TrainingCode1
Back to the Future: Towards Explainable Temporal Reasoning with Large Language ModelsCode1
Zero-Shot Compositional Policy Learning via Language GroundingCode1
Aya Dataset: An Open-Access Collection for Multilingual Instruction TuningCode1
Creative Agents: Empowering Agents with Imagination for Creative TasksCode1
Counterfactual Cycle-Consistent Learning for Instruction Following and Generation in Vision-Language NavigationCode1
LASeR: Learning to Adaptively Select Reward Models with Multi-Armed BanditsCode1
Show:102550
← PrevPage 10 of 46Next →

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