Papers with Code 아카이브
Instruction Following
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
- 하위 태스크
- 1
- 평가표
- 1
- 보존된 평가 기록
- 4
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.
코드
관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.
https://github.com/jacobandreas/instructions
공식 코드Alignment-based compositional semantics for instruction following
https://github.com/jacobkrantz/VLN-CE
공식 코드Waypoint Models for Instruction-guided Navigation in Continuous Environments
https://github.com/jamespark3922/localized-skd
공식 코드Localized Symbolic Knowledge Distillation for Visual Commonsense Models
https://github.com/jamespark3922/lskd
공식 코드Localized Symbolic Knowledge Distillation for Visual Commonsense Models
https://github.com/jchiyah/blockworld-repairs
공식 코드https://github.com/jcruan519/mm-camobj
공식 코드MM-CamObj: A Comprehensive Multimodal Dataset for Camouflaged Object Scenarios
https://github.com/jefferyzhan/griffon
공식 코드Griffon-G: Bridging Vision-Language and Vision-Centric Tasks via Large Multimodal Models
https://github.com/jeffhernandez1995/GenLlaVA
공식 코드https://github.com/jiachenwestlake/MMKD
공식 코드Adversarial Moment-Matching Distillation of Large Language Models
https://github.com/jiaxing-star/facialdynamic
공식 코드https://github.com/jihaonew/mm-instruct
공식 코드MM-Instruct: Generated Visual Instructions for Large Multimodal Model Alignment
https://github.com/jinhaoduan/sar
공식 코드https://github.com/jinhaoduan/shifting-attention-to-relevance
공식 코드https://github.com/jlin816/rewards-from-language
공식 코드https://github.com/joeyy5588/lacma
공식 코드LACMA: Language-Aligning Contrastive Learning with Meta-Actions for Embodied Instruction Following
https://github.com/john-hewitt/implicit-ins
공식 코드https://github.com/jonas-becker/problem-drift
공식 코드https://github.com/jongwooko/distillm
공식 코드DistiLLM: Towards Streamlined Distillation for Large Language Models
https://github.com/jongwooko/distillm-2
공식 코드DistiLLM-2: A Contrastive Approach Boosts the Distillation of LLMs
https://github.com/joonkeekim/Instructive-Decoding
공식 코드https://github.com/joshuaclymer/genies
공식 코드Generalization Analogies: A Testbed for Generalizing AI Oversight to Hard-To-Measure Domains
https://github.com/juliasun623/f-eval
공식 코드F-Eval: Assessing Fundamental Abilities with Refined Evaluation Methods
https://github.com/junjie-ye/muldimif
공식 코드https://github.com/juyongjiang/kasa
공식 코드KaSA: Knowledge-Aware Singular-Value Adaptation of Large Language Models
https://github.com/kaistai/flask
공식 코드FLASK: Fine-grained Language Model Evaluation based on Alignment Skill Sets
https://github.com/kaistai/instructir
공식 코드INSTRUCTIR: A Benchmark for Instruction Following of Information Retrieval Models
https://github.com/kehanlu/desta2.5-audio
공식 코드https://github.com/kehanlu/speech-ifeval
공식 코드https://github.com/keio-smilab23/dialmat
공식 코드DialMAT: Dialogue-Enabled Transformer with Moment-Based Adversarial Training
https://github.com/kexinhuang19/instructttseval
공식 코드https://github.com/kiddyboots216/lottery-ticket-adaptation
공식 코드Lottery Ticket Adaptation: Mitigating Destructive Interference in LLMs
https://github.com/kki2eve/agri-llava
공식 코드Agri-LLaVA: Knowledge-Infused Large Multimodal Assistant on Agricultural Pests and Diseases
https://github.com/kkk-an/ultraif
공식 코드https://github.com/klusai/tinyfabulist
공식 코드TF1-EN-3M: Three Million Synthetic Moral Fables for Training Small, Open Language Models
https://github.com/komeijiforce/cuckoo
공식 코드Cuckoo: An IE Free Rider Hatched by Massive Nutrition in LLM's Nest
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.