공대생 도전 일지

https://arxiv.org/abs/2309.13007 ReConcile: Round-Table Conference Improves Reasoning via Consensus among Diverse LLMsLarge Language Models (LLMs) still struggle with natural language reasoning tasks. Motivated by the society of minds (Minsky, 1988), we propose ReConcile, a multi-model multi-agent framework designed as a round table conference among diverse LLM agents. Rearxiv.org 기존 LLM은 새로운 생각..

https://arxiv.org/abs/2305.17144 Ghost in the Minecraft: Generally Capable Agents for Open-World Environments via Large Language Models with Text-based KnowledgeThe captivating realm of Minecraft has attracted substantial research interest in recent years, serving as a rich platform for developing intelligent agents capable of functioning in open-world environments. However, the current research..

https://arxiv.org/abs/2308.10144 ExpeL: LLM Agents Are Experiential LearnersThe recent surge in research interest in applying large language models (LLMs) to decision-making tasks has flourished by leveraging the extensive world knowledge embedded in LLMs. While there is a growing demand to tailor LLMs for custom decision-making tarxiv.org  이 논문도 LLM이 새로운 정보에 대해 어떻게 저장하거나 사용할지에 대한 논문입니다.문제: LLM을..

https://arxiv.org/abs/2305.16843 Randomized Positional Encodings Boost Length Generalization of TransformersTransformers have impressive generalization capabilities on tasks with a fixed context length. However, they fail to generalize to sequences of arbitrary length, even for seemingly simple tasks such as duplicating a string. Moreover, simply training on lonarxiv.org 학습 혹은 추론 때 토큰 길이에 대한 논문이..

https://arxiv.org/abs/2305.13172 Editing Large Language Models: Problems, Methods, and OpportunitiesDespite the ability to train capable LLMs, the methodology for maintaining their relevancy and rectifying errors remains elusive. To this end, the past few years have witnessed a surge in techniques for editing LLMs, the objective of which is to efficientlarxiv.org 이 논문은 새로운 정보를 반영하지 못 하는 LLM의 단점을..

https://arxiv.org/abs/2310.03214 FreshLLMs: Refreshing Large Language Models with Search Engine AugmentationMost large language models (LLMs) are trained once and never updated; thus, they lack the ability to dynamically adapt to our ever-changing world. In this work, we perform a detailed study of the factuality of LLM-generated text in the context of answeringarxiv.org검색을 통해 LLM의 최신 정보 미달을 해결하..

https://arxiv.org/abs/2302.04023 A Multitask, Multilingual, Multimodal Evaluation of ChatGPT on Reasoning, Hallucination, and InteractivityThis paper proposes a framework for quantitatively evaluating interactive LLMs such as ChatGPT using publicly available data sets. We carry out an extensive technical evaluation of ChatGPT using 23 data sets covering 8 different common NLP application taskarx..

https://arxiv.org/abs/2210.07128 Language Models of Code are Few-Shot Commonsense LearnersWe address the general task of structured commonsense reasoning: given a natural language input, the goal is to generate a graph such as an event -- or a reasoning-graph. To employ large language models (LMs) for this task, existing approaches ``serialize'arxiv.org이 모델은 명확하게 들어오지 않네요...?  COCOGEN은 구조적 상식(re..

https://arxiv.org/abs/1911.12543 How Can We Know What Language Models Know?Recent work has presented intriguing results examining the knowledge contained in language models (LM) by having the LM fill in the blanks of prompts such as "Obama is a _ by profession". These prompts are usually manually created, and quite possibly sub-oarxiv.orghttps://github.com/WooooDyy/LLM-Agent-Paper-List?tab=readm..

https://arxiv.org/abs/2303.08112 Eliciting Latent Predictions from Transformers with the Tuned LensWe analyze transformers from the perspective of iterative inference, seeking to understand how model predictions are refined layer by layer. To do so, we train an affine probe for each block in a frozen pretrained model, making it possible to decode everyarxiv.org 기존의 Logit Lens 방식은 Transformer의 출력..