Graham Neubig - Podcast Details

Shows

$Where What If Becomes What\'s Next$ Where What If Becomes What's NextFuture of AI: DeepSeek’s Transformative ImpactWhat if a small startup transformed the future of artificial intelligence in the blink of an algorithm?This conversation explores the transformative impact of DeepSeek, a groundbreaking AI chatbot that has disrupted the AI landscape by offering competitive performance at a fraction of the cost of its counterparts. Plus, DeepSeek’s open source model offers transparency and crowdsourced improvements while providing greater access to AI technology for developers and even nations. Experts discuss the innovations behind DeepSeek, its implications for the future of AI, and the broader trends and challenges in AI development, including its role in sci...2025-05-0730 min

PaperLedgeComputation and Language - TheAgentCompany Benchmarking LLM Agents on Consequential Real World TasksAlright Learning Crew, Ernis here, ready to dive into some seriously cool research! Today, we're talking about AI agents in the workplace. Now, we all use computers every day, right? Think about your work – how much of it is digital, done online? Well, researchers are wondering just how good AI is getting at actually doing some of that work for us. We've seen these amazing leaps in AI, especially with Large Language Models (LLMs). These aren't just chatbots anymore; they're AI agents that can interact with their environment, like browsing the web or even writing code. So, th...2025-05-0405 min

Practical AIOpen source AI to tackle your backlogVibe coding, agentic workflows, and AI-assisted pull requests? In this episode, Daniel and Chris chat with Robert Brennan and Graham Neubig of All Hands AI about how AI is transforming software development—from senior engineer productivity to open source agents that address GitHub issues. They dive into trust, tooling, collaboration, and what it means to build software in the era of AI agents. Whether you're coding from your laptop or your phone on a morning walk, the future is hands-free (and All Hands).Featuring:Robert Brennan – LinkedIn, XGraham Neubig – LinkedIn, XChris Benson – Website, GitHub, LinkedIn, XDaniel Whitenac...2025-04-1842 min

Daily Paper CastThe Danger of Overthinking: Examining the Reasoning-Action Dilemma in Agentic Tasks 🤗 Upvotes: 41 | cs.AI Authors: Alejandro Cuadron, Dacheng Li, Wenjie Ma, Xingyao Wang, Yichuan Wang, Siyuan Zhuang, Shu Liu, Luis Gaspar Schroeder, Tian Xia, Huanzhi Mao, Nicholas Thumiger, Aditya Desai, Ion Stoica, Ana Klimovic, Graham Neubig, Joseph E. Gonzalez Title: The Danger of Overthinking: Examining the Reasoning-Action Dilemma in Agentic Tasks Arxiv: http://arxiv.org/abs/2502.08235v1 Abstract: Large Reasoning Models (LRMs) represent a breakthrough in AI problem-solving capabilities, but their effectiveness in interactive environments can be limited. This paper introduces and analyzes overthinking in LRMs. A phenomenon whe...2025-02-1827 min

Daily Paper CastDemystifying Long Chain-of-Thought Reasoning in LLMs 🤗 Upvotes: 26 | cs.CL, cs.LG Authors: Edward Yeo, Yuxuan Tong, Morry Niu, Graham Neubig, Xiang Yue Title: Demystifying Long Chain-of-Thought Reasoning in LLMs Arxiv: http://arxiv.org/abs/2502.03373v1 Abstract: Scaling inference compute enhances reasoning in large language models (LLMs), with long chains-of-thought (CoTs) enabling strategies like backtracking and error correction. Reinforcement learning (RL) has emerged as a crucial method for developing these capabilities, yet the conditions under which long CoTs emerge remain unclear, and RL training requires careful design choices. In this study, we systematically investigate the...2025-02-0720 min

Daily Paper CastCowPilot: A Framework for Autonomous and Human-Agent Collaborative Web Navigation 🤗 Upvotes: 1 | cs.AI, cs.CL, cs.HC Authors: Faria Huq, Zora Zhiruo Wang, Frank F. Xu, Tianyue Ou, Shuyan Zhou, Jeffrey P. Bigham, Graham Neubig Title: CowPilot: A Framework for Autonomous and Human-Agent Collaborative Web Navigation Arxiv: http://arxiv.org/abs/2501.16609v1 Abstract: While much work on web agents emphasizes the promise of autonomously performing tasks on behalf of users, in reality, agents often fall short on complex tasks in real-world contexts and modeling user preference. This presents an opportunity for humans to collaborate with the agent and...2025-02-0121 min

ThursdAI - The top AI news from the past week📆 ThursdAI - Jan 16, 2025 - Hailuo 4M context LLM, SOTA TTS in browser, OpenHands interview & more AI newsHey everyone, Alex here 👋 Welcome back, to an absolute banger of a week in AI releases, highlighted with just massive Open Source AI push. We're talking a MASSIVE 4M context window context window model from Hailuo (remember when a jump from 4K to 16K seemed like a big deal?), a 8B omni model that lets you livestream video and glimpses of Agentic ChatGPT? This week's ThursdAI was jam-packed with so much open source goodness that the big companies were practically silent. But don't worry, we still managed to squeeze in some updates from OpenAI and Mis...2025-01-171h 40

Daily Paper CastAutoPresent: Designing Structured Visuals from Scratch 🤗 Upvotes: 7 | cs.CV, cs.CL Authors: Jiaxin Ge, Zora Zhiruo Wang, Xuhui Zhou, Yi-Hao Peng, Sanjay Subramanian, Qinyue Tan, Maarten Sap, Alane Suhr, Daniel Fried, Graham Neubig, Trevor Darrell Title: AutoPresent: Designing Structured Visuals from Scratch Arxiv: http://arxiv.org/abs/2501.00912v1 Abstract: Designing structured visuals such as presentation slides is essential for communicative needs, necessitating both content creation and visual planning skills. In this work, we tackle the challenge of automated slide generation, where models produce slide presentations from natural language (NL) instructions. We first introduce the...2025-01-0819 min

Tool Use - AI ConversationsHow to get #1 on SWE Bench (Ft Graham Neubig)Graham Neubig, the Chief Scientist and Co-Founder of All Hands AI joins us to talk about OpenHands and the future of coding agents Links https://www.all-hands.dev/ https://github.com/All-Hands-AI/OpenHands https://discord.gg/p2jcTRBTB8 Get the tools pip install tool-use-ai Connect with us https://x.com/ToolUseAI https://x.com/MikeBirdTech https://x.com/FieroTy https://x.com/gneubig 2025-01-0732 min

Daily Paper CastTraining Software Engineering Agents and Verifiers with SWE-Gym 🤗 Upvotes: 6 | cs.SE, cs.CL Authors: Jiayi Pan, Xingyao Wang, Graham Neubig, Navdeep Jaitly, Heng Ji, Alane Suhr, Yizhe Zhang Title: Training Software Engineering Agents and Verifiers with SWE-Gym Arxiv: http://arxiv.org/abs/2412.21139v1 Abstract: We present SWE-Gym, the first environment for training real-world software engineering (SWE) agents. SWE-Gym contains 2,438 real-world Python task instances, each comprising a codebase with an executable runtime environment, unit tests, and a task specified in natural language. We use SWE-Gym to train language model based SWE agents , achieving up to 19% absolute gai...2025-01-0126 min

ThursdAI - The top AI news from the past week📆 ThursdAI - Dec 26 - OpenAI o3 & o3 mini, DeepSeek v3 658B beating Claude, Qwen Visual Reasoning, Hume OCTAVE & more AI newsHey everyone, Alex here 👋I was hoping for a quiet holiday week, but whoa, while the last newsletter was only a week ago, what a looong week it has been, just Friday after the last newsletter, it felt like OpenAI has changed the world of AI once again with o3 and left everyone asking "was this AGI?" over the X-mas break (Hope Santa brought you some great gifts!) and then not to be outdone, DeepSeek open sourced basically a Claude 2.5 level behemoth DeepSeek v3 just this morning!Since the breaking news from DeepSeek took us by...2024-12-271h 35

Latent Space: The AI Engineer Podcast2024 in Agents [LS Live! @ NeurIPS 2024]Happy holidays! We’ll be sharing snippets from Latent Space LIVE! through the break bringing you the best of 2024! We want to express our deepest appreciation to event sponsors AWS, Daylight Computer, Thoth.ai, StrongCompute, Notable Capital, and most of all all our LS supporters who helped fund the gorgeous venue and A/V production!For NeurIPS last year we did our standard conference podcast coverage interviewing selected papers (that we have now also done for ICLR and ICML), however we felt that we could be doing more to help AI Engineers 1) get more industry-relevant content, and 2) re...2024-12-2548 min

Daily Paper CastTheAgentCompany: Benchmarking LLM Agents on Consequential Real World Tasks 🤗 Upvotes: 30 | cs.CL Authors: Frank F. Xu, Yufan Song, Boxuan Li, Yuxuan Tang, Kritanjali Jain, Mengxue Bao, Zora Z. Wang, Xuhui Zhou, Zhitong Guo, Murong Cao, Mingyang Yang, Hao Yang Lu, Amaad Martin, Zhe Su, Leander Maben, Raj Mehta, Wayne Chi, Lawrence Jang, Yiqing Xie, Shuyan Zhou, Graham Neubig Title: TheAgentCompany: Benchmarking LLM Agents on Consequential Real World Tasks Arxiv: http://arxiv.org/abs/2412.14161v1 Abstract: We interact with computers on an everyday basis, be it in everyday life or work, and many aspects of work can be...2024-12-2024 min

Daily Paper CastThe BrowserGym Ecosystem for Web Agent Research 🤗 Upvotes: 11 | cs.LG, cs.AI, cs.SE Authors: Thibault Le Sellier De Chezelles, Maxime Gasse, Alexandre Drouin, Massimo Caccia, Léo Boisvert, Megh Thakkar, Tom Marty, Rim Assouel, Sahar Omidi Shayegan, Lawrence Keunho Jang, Xing Han Lù, Ori Yoran, Dehan Kong, Frank F. Xu, Siva Reddy, Quentin Cappart, Graham Neubig, Ruslan Salakhutdinov, Nicolas Chapados, Alexandre Lacoste Title: The BrowserGym Ecosystem for Web Agent Research Arxiv: http://arxiv.org/abs/2412.05467v3 Abstract: The BrowserGym ecosystem addresses the growing need for efficient evaluation and benchmarking of web agents, particularly those...2024-12-1325 min

New Paradigm: AI Research SummariesCould Agent Workflow Memory Transform AI's Ability to Navigate and Solve Complex Web Tasks?This episode analyzes "Agent Workflow Memory," a study conducted by Zora Zhiruo Wang, Jiayuan Mao, Daniel Fried, and Graham Neubig from Carnegie Mellon University and the Massachusetts Institute of Technology. It explores the innovative approach of Agent Workflow Memory (AWM) in enhancing language model-based agents' ability to navigate and solve complex web tasks. The discussion delves into how AWM mimics human adaptability by learning and reusing task workflows from past experiences, thereby improving efficiency and success rates in both offline and online scenarios.The episode also reviews the empirical results from experiments conducted on the Mind2Web and...2024-12-1106 min

New Paradigm: AI Research SummariesHow Does AGORA BENCH Compare Language Models in Synthetic Data Generation?This episode analyzes the study "Evaluating Language Models as Synthetic Data Generators" by Seungone Kim, Juyoung Suk, Xiang Yue, Vijay Viswanathan, Seongyun Lee, Yizhong Wang, Kiril Gashteovski, Carolin Lawrence, Sean Welleck, and Graham Neubig, affiliated with institutions such as Carnegie Mellon University and KAIST AI. The discussion centers on the introduction of AGORA BENCH, a benchmark designed to assess the effectiveness of various language models in generating high-quality synthetic data.The episode delves into the comparative performance of six prominent language models, including GPT-4o and Claude-3.5-Sonnet, highlighting their distinct strengths in data generation tasks. It explores...2024-12-1005 min

Daily Paper CastMAmmoTH-VL: Eliciting Multimodal Reasoning with Instruction Tuning at Scale 🤗 Upvotes: 30 | cs.CL, cs.CV Authors: Jarvis Guo, Tuney Zheng, Yuelin Bai, Bo Li, Yubo Wang, King Zhu, Yizhi Li, Graham Neubig, Wenhu Chen, Xiang Yue Title: MAmmoTH-VL: Eliciting Multimodal Reasoning with Instruction Tuning at Scale Arxiv: http://arxiv.org/abs/2412.05237v1 Abstract: Open-source multimodal large language models (MLLMs) have shown significant potential in a broad range of multimodal tasks. However, their reasoning capabilities remain constrained by existing instruction-tuning datasets, which were predominately repurposed from academic datasets such as VQA, AI2D, and ChartQA. These datasets target sim...2024-12-1021 min

Daily Paper CastEvaluating Language Models as Synthetic Data Generators 🤗 Upvotes: 30 | cs.CL Authors: Seungone Kim, Juyoung Suk, Xiang Yue, Vijay Viswanathan, Seongyun Lee, Yizhong Wang, Kiril Gashteovski, Carolin Lawrence, Sean Welleck, Graham Neubig Title: Evaluating Language Models as Synthetic Data Generators Arxiv: http://arxiv.org/abs/2412.03679v1 Abstract: Given the increasing use of synthetic data in language model (LM) post-training, an LM's ability to generate high-quality data has become nearly as crucial as its ability to solve problems directly. While prior works have focused on developing effective data generation methods, they lack systematic comparison of different LMs...2024-12-0821 min

Daily Paper CastOpenScholar: Synthesizing Scientific Literature with Retrieval-augmented LMs 🤗 Paper Upvotes: 15 | cs.CL, cs.AI, cs.DL, cs.IR, cs.LG Authors: Akari Asai, Jacqueline He, Rulin Shao, Weijia Shi, Amanpreet Singh, Joseph Chee Chang, Kyle Lo, Luca Soldaini, Sergey Feldman, Mike D'arcy, David Wadden, Matt Latzke, Minyang Tian, Pan Ji, Shengyan Liu, Hao Tong, Bohao Wu, Yanyu Xiong, Luke Zettlemoyer, Graham Neubig, Dan Weld, Doug Downey, Wen-tau Yih, Pang Wei Koh, Hannaneh Hajishirzi Title: OpenScholar: Synthesizing Scientific Literature with Retrieval-augmented LMs Arxiv: http://arxiv.org/abs/2411.14199v1 Abstract: Scientific progress depends on researchers' ability to syn...2024-11-2323 min

Latent Space: The AI Engineer PodcastICLR 2024 — Best Papers & Talks (Benchmarks, Reasoning & Agents) — ft. Graham Neubig, Aman Sanger, Moritz Hardt)Our second wave of speakers for AI Engineer World’s Fair were announced! The conference sold out of Platinum/Gold/Silver sponsors and Early Bird tickets! See our Microsoft episode for more info and buy now with code LATENTSPACE.This episode is straightforwardly a part 2 to our ICLR 2024 Part 1 episode, so without further ado, we’ll just get right on with it!Timestamps[00:03:43] Section A: Code Edits and Sandboxes, OpenDevin, and Academia vs Industry — ft. Graham Neubig and Aman Sanger* [00:07:44] WebArena* [00:18:45] Sotopia* [00:24:00] Performance Improving Code Edits 2024-06-104h 29

ThursdAI - The top AI news from the past week📅 ThursdAI - May 9 - AlphaFold 3, im-a-good-gpt2-chatbot, Open Devin SOTA on SWE-Bench, DeepSeek V2 super cheap + interview with OpenUI creator & more AI newsHey 👋 (show notes and links a bit below)This week has been a great AI week, however, it does feel like a bit "quiet before the storm" with Google I/O on Tuesday next week (which I'll be covering from the ground in Shoreline!) and rumors that OpenAI is not just going to let Google have all the spotlight!Early this week, we got 2 new models on LMsys, im-a-good-gpt2-chatbot and im-also-a-good-gpt2-chatbot, and we've now confirmed that they are from OpenAI, and folks have been testing them with logic puzzles, role play and have bee...2024-05-101h 47

出埃及记ExodusS1E02｜Figure 01背后的具身智能：解析VLM、基础模型、硬件与交互【关于播客】欢迎大家一同加入出埃及记的旅程！关注底层逻辑，剖析前沿创新。出埃及记是一档从技术视角分析前沿科技进展的播客节目，我们会邀请全球顶尖院校的博士生或初创公司的创始人，一起聊聊人工智能等计算机科学或其他交叉领域的最新动态，分享他们的研究工作和相关洞察。欢迎大家订阅收听，和关注我们的同名微信公众号。;p【关于本期】2024年3月13日，Figure AI与Open AI合作推出的Figure 01人形机器人视频在社交媒体上热传。这段长达2分35秒的视频展示了机器人惊人的理解、判断、行动和自我评估能力。对比他们今年2月发布的机器人双腿行走和拿起塑料箱的视频，最新视频展示的人形机器人具有视觉能力并能表述所见画面，并采用端到端的神经网络，在接受人类的提问后，可以在“思索”2~3秒顺畅作答，手部动作的速度则接近人类。从OpenAI 宣布介入与之合作，到今天它们共同推出一个能够自主对话和决策的机器人，只有13天。在人工智能领域，国际上已有共识，即下一个重要挑战，就是实现具身智能（Embodied Intelligence），即能理解、推理并与物理世界互动的智能系统。全球范围内AI领域的巨头和风投机构的火力在今年也从软件卷入了门槛更高回报周期更长的硬件。我们也在这一期请到了机器人相关领域的三位博士生小伙伴，来跟我们分享机器人背后的视觉语言模型（VLM）、机器人领域的基础模型（foundation model）以及与之对应的精密的人形机器人都是什么？机器人与人的交互和“世界模型”又是怎样的？机器人精密控制的硬件是如何实现的？【嘉宾介绍】蔡易林（推特：@Yilin_Cai98）：佐治亚理工大学机器人学博士生，本硕毕业于上海交通大学和卡内基梅隆大学机器人研究所。研究聚焦于机器人灵巧操作、软体机器人和医疗机器人。江振宇（推特：@SteveTod1998）：德州大学奥斯汀分校计算机系博士生，师从 Yuke Zhu。本科毕业于清华大学电子系。研究聚焦于机器人视觉和机器人学习。目前在英伟达通用具身智能研究部门实习，参与英伟达人形机器人基础模型 GROOT 研发。朱昊（推特：@_Hao_Zhu）：卡内基梅隆大学语言技术研究所博士生，师从Graham Neubig 和 Yonatan Bisk。本科毕业于清华大学计算机系。研究聚焦于提升AI和人类在现实世界中的合作能力。【开降落伞】背景科普07:25 具身智能和非具身智能的区别？08:54 现在AI当中的主动交互是什么？具体是如何实现和环境之间交互的？09:51 机器人的发展有哪些阶段？11:14 机器人有哪些未来形态或者应用？除了机器人之外，具身智能还有哪些应用？15:06 具身智能的未来通向何方？视觉语言模型与基础模型18:33 如何让机器人可以理解图像和文字？21:09 什么是视觉语言模型的基石？22:33 怎样看待Figure 01背后的VLM?23:30 3D视觉近期在做什么？27:09 为什么需要人形机器人？人形机器人和其他的embodiment之间存在一些差异吗？这些差异是什么？32:07 机器人策略的最终形态会是怎样？33:00 机器人的基础模型的输入输出模态是什么样的？34:34 机器人基础模型的数据通过什么方式收集？一个具体的数据飞轮可能是什么样的？模拟交互环境、人机交互与世界模型37:36 模拟交互环境的优点?41:13 现有交互环境的类型和评价体系?45:59 在机器人和人的交互当中，衡量方式是什么，难点又是什么？49:12 世界模型是什么？51:00 世界模型可以怎样帮助具身智能？机器人硬件52:14 人形机器人的结构设计、硬件是怎样的？应该包含哪些模块？每个模块的特点，设计困难和对应的解决方案都有哪些？1:00:51 其他机器人形态的传统控制方法有哪些？Embodied AI/是如何参与到对于其他机器人形态的控制当中的呢？1:04:34 传统机器人的控制架构在做什么？1:08:04 大家接下来的研究方向？怎样看待具身智能和AGI？【关于主播】刘杰尼：社科传媒本计算机研，Ex高盛TMT投行吗喽，接触过各种行业和国内外项目，混过大厂小厂和创业公司，并有幸（有可能是不幸）经历中国上一个TMT时代的完整兴衰，试图更好地活在科技与人文的交叉点。徐胖虎：电子工程本，计算机硕博，现加州大学某海景分校论文力工，incoming Meta 机器学习实习生。研究经历覆盖LLMs、类脑神经网络和软硬件效率计算，试图理解生物人脑和暴力美学的大模型之间的共通之处。【音乐版权】 Time by Pold soundcloud.com｜Free Download / Stream: tinyurl.com｜Promoted by Audio Library tinyurl.com Look At The Clouds by Declan DP｜Free Download / Stream: bit.ly｜Promoted by Audio Library bit.ly 2024-04-141h 15

SlatorPod#175 Where DeepL Beats ChatGPT with Graham NeubigIn this week’s SlatorPod, we are joined by Graham Neubig, Associate Professor of Computer Science at Carnegie Mellon University, to discuss his research on multilingual natural language processing (NLP) and machine translation (MT). Graham discusses the research at Neulab, where they focus on various areas of NLP, including incorporating broad knowledge bases into NLP models and code generation.Graham expands on his Zeno GPT-MT Report comparing large language models (LLMs) with special-purpose machine translation models like Google Translate, Microsoft Translate, and DeepL. He revealed that GPT-4 was competitive from English to other languages, but st...2023-07-1438 min

CS224USasha Rush on NLP research, engineering, and educationCoding puzzles, practices, and education, structured prediction, the culture of Hugging Face, large models, and the energy of New York. Transcript: https://web.stanford.edu/class/cs224u/podcast/rush/ Sasha's website Sasha on Twitter Sasha on the Humans of AI podcast Sasha on The Thesis Review Podcast with Sean Welleck Sasha on the Talking Machines Podcast Sasha interviewed by Sayak Paul Hugging Face PyTorch The Annotated Transformer The Annotated Alice The Annotated S4 Sasha and Dan Oneață's declarative graphics library Chalk Drawing Big Ben in Chalk OpenNMT Ken Shan Blog post by Ken and Dy...2022-10-041h 22

The Thesis Review[22] Graham Neubig - Unsupervised Learning of Lexical InformationGraham Neubig is an Associate Professor at Carnegie Mellon University. His research focuses on language and its role in human communication, with the goal of breaking down barriers in human-human or human-machine communication through the development of NLP technologies. Graham’s PhD thesis is titled "Unsupervised Learning of Lexical Information for Language Processing Systems", which he completed in 2012 at Kyoto University. We discuss his PhD work related to the fundamental processing units that NLP systems use to process text, including non-parametric Bayesian models, segmentation, and alignment problems, and discuss how his perspective on machine translation has evolved over time. Episode no...2021-04-021h 02

Leading NLP Ninjaep27: 今年度の振り返りとこれからについて第２７回では，２０１８年度の振り返りの特別回として，アンケート結果の共有，NLPのおすすめ勉強方法と教材，論文の読み方，期末の振り返りと今後について話しました．アンケートの振り返り NLPの勉強の仕方，論文の読み方今期，来期のはなし紹介した書籍やサイト自然言語処理の基本と技術 Graham Neubig先生のNLPプログラミングチュートリアル言語処理のための機械学習入門番組への支援は，こちらからお待ちしております． --- Support this podcast: https://podcasters.spotify.com/pod/show/lnlp-ninja/support 2019-03-3117 min

NLP Highlights63 - Neural Lattice Language Models, with Jacob BuckmanTACL 2018 paper by Jacob Buckman and Graham Neubig. Jacob tells us about marginalizing over latent structure in a sentence by doing a clever parameterization of a lattice with a model kind of like a tree LSTM. This lets you treat collocations as multi-word units, or allow words to have multiple senses, without having to commit to a particular segmentation or word sense disambiguation up front. We talk about how this works and what comes out. One interesting result that comes out of the sense lattice: learning word senses from a language modeling objective tends to give you senses that capture...2018-08-0230 min

NLP Highlights29 - Neural machine translation via binary code prediction, with Graham NeubigACL 2017 paper, by Yusuke Oda and others (including Graham Neubig) at Nara Institute of Science and Technology (Graham is now at Carnegie Mellon University). Graham comes on to talk to us about neural machine translation generally, and about this ACL paper in particular. We spend the first half of the episode talking about major milestones in neural machine translation and why it is so much more effective than previous methods (spoiler: stronger language models help a lot). We then talk about the specifics of binary code prediction, how it's related to a hierarchical or class-factored softmax, and how to make...2017-07-1438 min