PODCAST

Seventy3

任雨山

73播客，名字取材于Sheldon最喜欢的数字，内容由NotebookLM生成，每天跟随AI读AI业界论文。

technology

All Episodes

00:22:17

【第113期】ASAL：使用LLM自动搜索人工生命

Seventy3 ·

2025/01/21

Seventy3: 用NotebookLM将论文生成播客，让大家跟着AI一起进步。今天的主题是：Automating the Search for Artificial Life with Foundation ModelsSummaryThis research paper introduces Automated Search for Artificial Life (ASAL), a novel method using foundation models (FMs) to automate the discovery of interesting artificial life (ALife) simulations. ASAL employs FMs to evaluate simulations across diverse substrates (like Boids and Lenia), enabling three search strategies: supervised target searching, open-endedness searching, and

zh-cn

00:23:54

【第112期】Differentiable Cache Augmentation

Seventy3 ·

2025/01/20

Seventy3: 用NotebookLM将论文生成播客，让大家跟着AI一起进步。今天的主题是：Deliberation in Latent Space via Differentiable Cache AugmentationSummaryThis research paper explores a novel method for improving large language models (LLMs) by augmenting their internal cache with latent embeddings generated by a separate "coprocessor" model. This coprocessor, trained using standard language modeling techniques on a large dataset, learns to distill additional computation into the LLM's cache, enhancing its reasoning abilities

zh-cn

00:21:46

【第111期】LearnLM：Gemini在教育场景的应用

Seventy3 ·

2025/01/19

Seventy3: 用NotebookLM将论文生成播客，让大家跟着AI一起进步。今天的主题是：LearnLM: Improving Gemini for LearningSummaryThis research paper details the development and evaluation of LearnLM, a Google AI model designed for educational applications. LearnLM improves upon existing models by incorporating "pedagogical instruction following," allowing developers to specify desired teaching behaviors through system-level instructions. Extensive human evaluations, involving pedagogy experts, demonstrated LearnLM's superior

zh-cn

00:20:25

【第110期】PC Agent：通过学习人类认知过程来执行复杂的数字化工作

Seventy3 ·

2025/01/18

Seventy3: 用NotebookLM将论文生成播客，让大家跟着AI一起进步。今天的主题是：PC Agent: While You Sleep, AI Works -- A Cognitive Journey into Digital WorldSummaryThis research paper introduces PC Agent, a novel AI system designed to perform complex digital work by learning from human cognitive processes. The system comprises three key components: PC Tracker for data collection, a cognition completion pipeline for data refinement and semantic understanding, and a multi-agent system for task execution. PC Agent demonstrates

zh-cn

00:16:13

【第109期】AutoFeedback：使用智能体做自动反馈系统

Seventy3 ·

2025/01/17

Seventy3: 用NotebookLM将论文生成播客，让大家跟着AI一起进步。今天的主题是：Using Generative AI and Multi-Agents to Provide Automatic FeedbackSummaryThis research paper explores using a multi-agent system called AutoFeedback to improve the quality of automatically generated feedback for student responses in science assessments. AutoFeedback uses two AI agents: one to generate initial feedback and another to validate and refine it, addressing common issues like over-praise and over-inference found in single-agent large language

zh-cn

00:14:21

【第108期】PAE：能够自主学习新的网页导航技能

Seventy3 ·

2025/01/16

Seventy3: 用NotebookLM将论文生成播客，让大家跟着AI一起进步。今天的主题是：Proposer-Agent-Evaluator(PAE): Autonomous Skill Discovery For Foundation Model Internet AgentsSummaryThis research introduces Proposer-Agent-Evaluator (PAE), a novel system that enables foundation model agents to autonomously learn new skills for web navigation. PAE leverages a context-aware task proposer to suggest tasks, an agent policy to execute them, and an autonomous evaluator to provide feedback via reinforcement learning. Experiments on

zh-cn

00:18:26

【第107期】SGD-SaI：替代Adam类优化方法

Seventy3 ·

2025/01/15

Seventy3: 用NotebookLM将论文生成播客，让大家跟着AI一起进步。今天的主题是：No More Adam: Learning Rate Scaling at Initialization is All You NeedSummaryThe research introduces SGD-SaI, a novel optimization method that significantly improves the memory efficiency and training speed of large neural networks. Unlike adaptive methods like AdamW, SGD-SaI scales learning rates at initialization based on gradient signal-to-noise ratios, eliminating the need for storing and updating second-order momentum. This approach achieves

zh-cn

00:12:11

【第106期】ScaleOT：保护隐私的大型语言模型离站微调的新型框架

Seventy3 ·

2025/01/14

Seventy3: 用NotebookLM将论文生成播客，让大家跟着AI一起进步。今天的主题是：ScaleOT: Privacy-utility-scalable Offsite-tuning with Dynamic LayerReplace and Selective Rank CompressionSummaryThis research introduces ScaleOT, a novel framework for privacy-preserving offsite tuning of large language models (LLMs). ScaleOT addresses limitations of existing methods by using reinforcement learning to determine layer importance, replacing less important layers with lightweight networks ("harmonizers"), and employing rank reduction to

zh-cn

00:21:27

【第105期】MAXINFORL：最大化对底层任务信息增益的强化学习

Seventy3 ·

2025/01/13

Seventy3: 用NotebookLM将论文生成播客，让大家跟着AI一起进步。今天的主题是：MaxInfoRL: Boosting exploration in reinforcement learning through information gain maximizationSummaryThe paper introduces MAXINFORL, a novel reinforcement learning (RL) framework that improves exploration by maximizing information gain about the underlying task. It augments existing off-policy RL methods with directed exploration, using intrinsic rewards derived from model epistemic uncertainty to guide exploration more effectively than standard

zh-cn

00:15:47

【第104期】STAR：无梯度的进化优化算法

Seventy3 ·

2025/01/12

Seventy3: 用NotebookLM将论文生成播客，让大家跟着AI一起进步。今天的主题是：STAR: Synthesis of Tailored ArchitecturesSummaryThis research paper introduces STAR, a novel framework for automated deep learning architecture synthesis. STAR utilizes a hierarchical search space based on linear input-varying systems, numerically encoded as "genomes," which are optimized using gradient-free evolutionary algorithms. The system is evaluated on autoregressive language modeling, demonstrating significant improvements in model quality,

zh-cn

00:21:55

【第103期】开源和闭源大型语言模型的比较研究

Seventy3 ·

2025/01/11

Seventy3: 用NotebookLM将论文生成播客，让大家跟着AI一起进步。今天的主题是：The Open Source Advantage in Large Language Models (LLMs)SummaryThis research paper compares open-source and closed-source Large Language Models (LLMs), examining their development, performance, accessibility, and ethical implications. Open-source LLMs, like LLaMA and BLOOM, prioritize accessibility and community collaboration, while closed-source models, such as GPT-4, excel in performance due to proprietary data and resources. The paper analyzes the

zh-cn

00:17:14

【第102期】Byte Latent Transformer (BLT)：用byte级替代token级

Seventy3 ·

2025/01/10

Seventy3: 用NotebookLM将论文生成播客，让大家跟着AI一起进步。今天的主题是：Byte Latent Transformer: Patches Scale Better Than TokensSummaryThe paper introduces the Byte Latent Transformer (BLT), a novel large language model architecture that processes raw byte data without tokenization. BLT dynamically groups bytes into patches based on entropy, allocating computational resources efficiently. Experimental results demonstrate BLT's competitive performance with tokenization-based models, particularly showcasing improved

zh-cn

00:12:30

【第101期】Large Concept Models (LCMs)

Seventy3 ·

2025/01/09

Seventy3: 用NotebookLM将论文生成播客，让大家跟着AI一起进步。今天的主题是：Large Concept Models: Language Modeling in a Sentence Representation SpaceSummaryThis research paper introduces Large Concept Models (LCMs), a novel approach to language modeling that operates on sentence embeddings instead of individual tokens. LCMs aim to mimic human-like abstract reasoning by processing higher-level semantic representations, improving long-form text generation and zero-shot cross-lingual performance. The authors explore various LCM

zh-cn

00:17:43

【第100期】SLM更懂LLM提示词

Seventy3 ·

2025/01/08

Seventy3: 用NotebookLM将论文生成播客，让大家跟着AI一起进步。今天的主题是：Smaller Language Models Are Better Instruction EvolversSummaryThis research paper investigates the surprising effectiveness of smaller language models (SLMs) in improving instruction data for larger language models (LLMs). The authors challenge the common assumption that larger models are always superior for this task, demonstrating through experiments across three scenarios that SLMs generate more complex and diverse instructions. They attribute this

zh-cn

00:10:25

【第99期】GREATER：一种对于小模型的提示词优化技术

Seventy3 ·

2025/01/07

Seventy3: 用NotebookLM将论文生成播客，让大家跟着AI一起进步。今天的主题是：GReaTer: Gradients over Reasoning Makes Smaller Language Models Strong Prompt OptimizersSummaryThe paper introduces GREATER, a novel prompt optimization technique for smaller language models. Unlike existing methods that rely on large, expensive LLMs for feedback, GREATER uses gradient information directly from the task loss to refine prompts. This allows smaller models to achieve performance comparable to or exceeding that of larger models on various

zh-cn

00:31:13

【第98期】SPaR：通过搜索树改进LLM指令遵循

Seventy3 ·

2025/01/06

Seventy3: 用NotebookLM将论文生成播客，让大家跟着AI一起进步。今天的主题是：SPaR: Self-Play with Tree-Search Refinement to Improve Instruction-Following in Large Language ModelsSummaryThis research introduces SPAR, a self-play framework using tree-search refinement to improve instruction-following in large language models (LLMs). SPAR addresses the limitations of existing methods by generating comparable preference pairs free from irrelevant variations, focusing on key differences crucial for successful instruction-following

zh-cn

00:28:15

【第97期】SCBench：基于KV Cache的评估长上下文LLM基准

Seventy3 ·

2025/01/05

Seventy3: 用NotebookLM将论文生成播客，让大家跟着AI一起进步。今天的主题是：SCBench: A KV Cache-Centric Analysis of Long-Context MethodsSummaryThe paper introduces SCBench, a new benchmark for evaluating long-context Large Language Models (LLMs). SCBench focuses on the key role of the KV cache in LLM inference, analyzing its lifecycle across multiple requests and shared contexts. The benchmark assesses four key long-context abilities through twelve tasks, testing various long-context methods on multiple open-source LLMs

zh-cn

00:16:13

【第96期】AsyncLM：异步LLM函数调用

Seventy3 ·

2025/01/04

Seventy3: 用NotebookLM将论文生成播客，让大家跟着AI一起进步。今天的主题是：Asynchronous LLM Function CallingSummaryThis research paper introduces AsyncLM, a system designed to enhance the efficiency of Large Language Models (LLMs) by enabling asynchronous function calls. Unlike current synchronous methods where LLMs block while awaiting function execution, AsyncLM allows concurrent operation, significantly reducing task completion latency. This is achieved through an interrupt mechanism that notifies the LLM when functions

zh-cn

00:15:03

【第95期】Student-Informed Teacher Training

Seventy3 ·

2025/01/03

Seventy3: 用NotebookLM将论文生成播客，让大家跟着AI一起进步。今天的主题是：Student-Informed Teacher TrainingSummaryThis research introduces a novel framework for imitation learning that addresses the challenge of teacher-student asymmetry. The method jointly trains a teacher and student policy, where the teacher learns behaviors easily imitated by the student despite the student's limited observability. This is achieved by adding a penalty term to the teacher's reward function and incorporating a supervised alignment step

zh-cn

00:18:23

【第94期】AgentTrek：为GUI Agent生成高质量数据的pipeline

Seventy3 ·

2025/01/02

Seventy3: 用NotebookLM将论文生成播客，让大家跟着AI一起进步。今天的主题是：AgentTrek: Agent Trajectory Synthesis via Guiding Replay with Web TutorialsSummaryThe paper introduces AgentTrek, a novel pipeline for synthesizing high-quality training data for Graphical User Interface (GUI) agents. AgentTrek leverages web tutorials to generate large-scale, multi-step agent trajectories, significantly reducing the cost and effort compared to human annotation. The pipeline automatically gathers and processes tutorials, uses a

zh-cn

578 results

Seventy3

All Episodes

Similar Podcasts

Lead With AI

Tamara Nall

Linux in the Ham Shack

Black Sparrow Media

Data Engineering Podcast

Tobias Macey

Eye On A.I.

Craig S. Smith

Reflections

Livewire Labs