공대생 도전 일지

https://arxiv.org/abs/2305.10142 Improving Language Model Negotiation with Self-Play and In-Context Learning from AI FeedbackWe study whether multiple large language models (LLMs) can autonomously improve each other in a negotiation game by playing, reflecting, and criticizing. We are interested in this question because if LLMs were able to improve each other, it would imply thearxiv.org 음 여기선 이..

https://arxiv.org/abs/2308.08155 AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent ConversationAutoGen is an open-source framework that allows developers to build LLM applications via multiple agents that can converse with each other to accomplish tasks. AutoGen agents are customizable, conversable, and can operate in various modes that employ combiarxiv.org 이건 아래 논문과 엄청 비슷한 느낌이네요2024...

https://arxiv.org/abs/1706.05125 Deal or No Deal? End-to-End Learning for Negotiation DialoguesMuch of human dialogue occurs in semi-cooperative settings, where agents with different goals attempt to agree on common decisions. Negotiations require complex communication and reasoning skills, but success is easy to measure, making this an interestingarxiv.org 아직 Transformer 구조도 없어 RNN기반의 GRU 모델을 통..

https://arxiv.org/abs/2412.04139 Monet: Mixture of Monosemantic Experts for TransformersUnderstanding the internal computations of large language models (LLMs) is crucial for aligning them with human values and preventing undesirable behaviors like toxic content generation. However, mechanistic interpretability is hindered by polysemanticityarxiv.org 처음에는 단순 MOE에 SAE를 붙인 줄 알았는데 MOE를 최대한 발전시켜 추가 ..

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4828339 The Advent of the AI Negotiator: Negotiation Dynamics in the Age of Smart AlgorithmsThe Advent of the AI Negotiator: Negotiation Dynamics in the Age of Smart Algorithms 26 Pages Posted: 15 May 2024 Date Written: May 15, 2024 Abstract Artificial Intelligence (AI) applications are increasingly used in negotiations. In this essay, I invest..

https://www.nber.org/papers/w32685 Demand for Artificial Intelligence in Settlement NegotiationsJoshua Gans has drawn on the findings of his research for both compensated speaking engagements and consulting engagements. He has written the books Prediction Machines, Power & Prediction, and Innovation + Equality on the economics of AI for which he recewww.nber.org엥 제가 생각했던 논문이랑은 좀 다른 방향의 논문이었네요협상 ..

https://www.nature.com/articles/s41467-022-34473-5 저는 자연어를 통한 협상을 기대했드만 그냥 강화학습 이네요제한된 액션 공간에서 Agent끼리 서로 협상하고, 협상 내용에 대해 자신에게 가장 유리한 결과를 가져오는 협상을 진행하는 그런 내용이었습니다.학습은 강화학습을 통해 진행되었고요 그래도 협약을 깰 수 있는 존재를 만들어 보고, 그런 존재가 계속 이기다보니 새로운 역할을 하는 존재가 등장하며 협약을 깰 수 있는 존재가 결과적으로 이긴다고 해도, 대부분의 협약을 지키면서 게임을 진행하는 것이 인상적이긴 하네요  논문 제목Negotiation and Honesty in Artificial Intelligence Methods for the Board Game of..

https://arxiv.org/abs/2401.03568 Agent AI: Surveying the Horizons of Multimodal InteractionMulti-modal AI systems will likely become a ubiquitous presence in our everyday lives. A promising approach to making these systems more interactive is to embody them as agents within physical and virtual environments. At present, systems leverage existingarxiv.org 이 논문은 LLM 뿐많이 아니라 VLM을 활용하여 AGI 도달하기 위한 A..

https://arxiv.org/abs/2308.01552 InterAct: Exploring the Potentials of ChatGPT as a Cooperative AgentThis research paper delves into the integration of OpenAI's ChatGPT into embodied agent systems, evaluating its influence on interactive decision-making benchmark. Drawing a parallel to the concept of people assuming roles according to their unique strengtarxiv.org ReAct를 발전시킨 논문이네요2024.11.26 - [..

https://arxiv.org/abs/2402.05863 How Well Can LLMs Negotiate? NegotiationArena Platform and AnalysisNegotiation is the basis of social interactions; humans negotiate everything from the price of cars to how to share common resources. With rapidly growing interest in using large language models (LLMs) to act as agents on behalf of human users, such LLM agarxiv.org 음 이 논문은 그래도 LLM까지는 갔지만 LLM을 학습하거..