m0o0scar

[github] ai/nanoid

Source

JavaScript / 1.8K lines of code. A tiny (124 bytes), secure, URL-friendly, unique string ID generator for JavaScript

URL: https://github.com/ai/nanoid

Conversation

m0o0scar

[arxiv] LazyLLM: Dynamic Token Pruning for Efficient Long Context LLM Inference

Source

Authors: Qichen Fu, Minsik Cho, Thomas Merth, Sachin Mehta, Mohammad Rastegari, Mahyar Najibi

Published on: 19 Jul 2024

Abstract: The inference of transformer-based large language models consists of two sequential stages: 1) a prefilling stage to compute the KV cache of prompts and generate the first token, and 2) a decoding stage to generate subsequent tokens. For long prompts, the KV cache must be computed for all tokens during the prefilling stage, which can significantly increase the time needed to generate the first token. Consequently, the prefilling stage may become a bottleneck in the generation process. An open question remains whether all prompt tokens are essential for generating the first token. To answer this, we introduce a novel method, LazyLLM, that selectively computes the KV for tokens important for the next token prediction in both the prefilling and decoding stages. Contrary to static pruning

m0o0scar

[github] THUDM/CogVideo

Source

Python / 18.0K lines of code. Text-to-video generation: CogVideoX (2024) and CogVideo (ICLR 2023)

URL: https://github.com/THUDM/CogVideo?tab=readme-ov-file

Conversation

m0o0scar

[arxiv] Very Large-Scale Multi-Agent Simulation in AgentScope

Source

Authors: Xuchen Pan, Dawei Gao, Yuexiang Xie, Zhewei Wei, Yaliang Li, Bolin Ding, Ji-Rong Wen, Jingren Zhou

Published on: 25 Jul 2024

Abstract: Recent advances in large language models (LLMs) have opened new avenues for applying multi-agent systems in very large-scale simulations. However, there remain several challenges when conducting multi-agent simulations with existing platforms, such as limited scalability and low efficiency, unsatisfied agent diversity, and effort-intensive management processes. To address these challenges, we develop several new features and components for AgentScope, a user-friendly multi-agent platform, enhancing its convenience and flexibility for supporting very large-scale multi-agent simulations. Specifically, we propose an actor-based distributed mechanism as the underlying technological infrastructure towards great scalability and high efficiency, and provide flexible environment support for simul

m0o0scar

[arxiv] Self-Taught Evaluators

Source

Authors: Tianlu Wang, Ilia Kulikov, Olga Golovneva, Ping Yu, Weizhe Yuan, Jane Dwivedi-Yu, Richard Yuanzhe Pang, Maryam Fazel-Zarandi, Jason Weston, Xian Li

Published on: 5 Aug 2024

Abstract: Model-based evaluation is at the heart of successful model development -- as a reward model for training, and as a replacement for human evaluation. To train such evaluators, the standard approach is to collect a large amount of human preference judgments over model responses, which is costly and the data becomes stale as models improve. In this work, we present an approach that aims to im-prove evaluators without human annotations, using synthetic training data only. Starting from unlabeled instructions, our iterative self-improvement scheme generates contrasting model outputs and trains an LLM-as-a-Judge to produce reasoning traces and final judgments, repeating this training at each new iteration using the improved predictions. Without any labeled preference data, our

m0o0scar

[github] modelscope/agentscope

Source

Python / 40.3K lines of code. Start building LLM-empowered multi-agent applications in an easier way.

URL: https://github.com/modelscope/agentscope

Conversation

m0o0scar

[github] microsoft/qlib

Source

Python / 47.2K lines of code. Qlib is an AI-oriented quantitative investment platform that aims to realize the potential, empower research, and create value using AI technologies in quantitative investment, from exploring ideas to implementing productions. Qlib supports diverse machine learning modeling paradigms. including supervised learning, market dynamics modeling, and RL.

URL: https://github.com/microsoft/qlib

Conversation

m0o0scar

[github] microsoft/qlib

Source

Python / 47.2K lines of code. Qlib is an AI-oriented quantitative investment platform that aims to realize the potential, empower research, and create value using AI technologies in quantitative investment, from exploring ideas to implementing productions. Qlib supports diverse machine learning modeling paradigms. including supervised learning, market dynamics modeling, and RL.

URL: https://github.com/microsoft/qlib

Conversation

m0o0scar

[arxiv] Reflexion: Language Agents with Verbal Reinforcement Learning

Source

Authors: Noah Shinn, Federico Cassano, Edward Berman, Ashwin Gopinath, Karthik Narasimhan, Shunyu Yao

Abstract: Large language models (LLMs) have been increasingly used to interact with external environments (e.g., games, compilers, APIs) as goal-driven agents. However, it remains challenging for these language agents to quickly and efficiently learn from trial-and-error as traditional reinforcement learning methods require extensive training samples and expensive model fine-tuning. We propose Reflexion, a novel framework to reinforce language agents not by updating weights, but instead through linguistic feedback. Concretely, Reflexion agents verbally reflect on task feedback signals, then maintain their own reflective text in an episodic memory buffer to induce better decision-making in subsequent trials. Reflexion is flexible enough to incorporate various types (scalar values or free-form language) and sources (externa

m0o0scar

[arxiv] Very Large-Scale Multi-Agent Simulation in AgentScope

Source

Authors: Xuchen Pan, Dawei Gao, Yuexiang Xie, Zhewei Wei, Yaliang Li, Bolin Ding, Ji-Rong Wen, Jingren Zhou

Published on: 25 Jul 2024

Abstract: Recent advances in large language models (LLMs) have opened new avenues for applying multi-agent systems in very large-scale simulations. However, there remain several challenges when conducting multi-agent simulations with existing platforms, such as limited scalability and low efficiency, unsatisfied agent diversity, and effort-intensive management processes. To address these challenges, we develop several new features and components for AgentScope, a user-friendly multi-agent platform, enhancing its convenience and flexibility for supporting very large-scale multi-agent simulations. Specifically, we propose an actor-based distributed mechanism as the underlying technological infrastructure towards great scalability and high efficiency, and provide flexible environment support for simu