Aller au contenu
Global sound
Learning GenAI via SOTA Papers

Learning GenAI via SOTA Papers

de Yun Wu

Dernier épisode
EP540: How AI Robots Dream Before Moving

À propos de ce podcast

This podcast is focusing on sharing the papers on GenAI related topic, especially the SOTA (State of the Art) papers that are the foundations of GenAI work. It shows how these researches paved the way to the GenAI tools that we are using every day such as ChatGPT, Gemini, Claude Code etc.

Épisodes

Les 20 épisodes les plus récents sur 540

  • EP540: How AI Robots Dream Before Moving · 22 min · Saison 1 Épisode 540Title: PreAct-Nav: Agentic Reasoning Before Action for Urban NavigationSource: http://arxiv.org/abs/2610.04916v1 Summary: This work establishes a foundational paradigm shift for…
  • EP539: How GitSwarm Cures AI Amnesia · 22 min · Saison 1 Épisode 539Title: GitSwarm: Decentralized Compounding InferenceSource: http://arxiv.org/abs/2610.04862v1 Summary: This paper introduces a decentralized compounding inference framework that…
  • EP538: How DAGO Slashes AI Workflow Costs · 21 min · Saison 1 Épisode 538Title: Fusion is the New Mutation: Bandit-Guided Evolution on Workflow GraphsSource: http://arxiv.org/abs/2610.05284v1 Summary:This paper introduces a foundational framework for…
  • EP537: AI Rewrote Its Engine to Cheat Benchmarks · 19 min · Saison 1 Épisode 537Title: SEIS: Self-Evolving Inference Systems Source: http://arxiv.org/abs/2610.04646v1 Summary: This work introduces a foundational paradigm for self-evolving inference systems…
  • EP536: Why AI Benchmarks Hide Agent Failures · 24 min · Saison 1 Épisode 536Title: Agentic Cognitive Depth: Operational Criteria for Evaluating LLM Agents Source: http://arxiv.org/abs/2610.04168v1 Summary:This paper establishes foundational operational…
  • EP535: AI Counselors That Evolve in Real Time · 24 min · Saison 1 Épisode 535Title: PsyEvo: A Personalized Counseling Agent That Self-Evolves at Test Time Source: http://arxiv.org/abs/2610.02885v1 Summary: This paper presents a foundational approach for…
  • EP534: Robots That Learn to Fix Their Mistakes · 22 min · Saison 1 Épisode 534Title: MobiAgent: Dual-Loop Recursive Policy Self-Improvement for Long-Horizon Mobile ManipulationSource: http://arxiv.org/abs/2610.03476v1 Summary: This paper introduces a…
  • EP533: Why Smarter AI Invents Alien Languages · 20 min · Saison 1 Épisode 533Title: On Language Drift during RLVR Post-Training Source: http://arxiv.org/abs/2610.02015v1 Summary: This paper investigates key post-training dynamics within Reinforcement…
  • EP532: MemFit Reinvents AI Agent Memory · 20 min · Saison 1 Épisode 532Title: MemFit: Efficient Long-Term Agentic Memory Source: http://arxiv.org/abs/2610.00872v1 Summary: This paper addresses a core structural bottleneck in agentic AI by proposing…
  • EP531: Fixing multi-step AI without expensive critics · 18 min · Saison 1 Épisode 531Title: T2SPO: Trajectory-to-Step Policy Optimization for Agentic Reinforcement LearningSource: http://arxiv.org/abs/2610.00388v1 Summary:This paper presents a novel policy…
  • EP530: JevSpawn Solves the AI Token Bottleneck · 23 min · Saison 1 Épisode 530Title: JevSpawn: Adaptive Agentic Inference through Compositional Action SpacesSource: http://arxiv.org/abs/2610.00437v1 Summary:This paper proposes a foundational agentic…
  • EP529: Single AI Agents Beat Complex Frameworks · 22 min · Saison 1 Épisode 529Title: How Much of a Harness Does a Strong Agent Need for Autonomous ML Engineering? Source: http://arxiv.org/abs/2609.40303v1Summary: This work provides a critical evaluation…
  • EP528: How Hindsight Makes Small AI Beat GPT · 22 min · Saison 1 Épisode 528Title: Beyond Prediction: Steering VLM Agents with Retrospective World ModelingSource: http://arxiv.org/abs/2609.39101v1 Summary:This paper introduces a foundational framework…
  • EP527: How Confidence Filtering Fixes AI Reasoning · 22 min · Saison 1 Épisode 527Title: Provable Test-Time Scaling for Beam Search in LLM Reasoning Source: http://arxiv.org/abs/2609.38672v1 Summary: This work establishes theoretical and algorithmic…
  • EP526: From Model Distillation to Agent Distillation · 23 min · Saison 1 Épisode 526Title: Distilling Agentic Systems: A Roadmap across Models, Artifacts, and HarnessesSource: http://arxiv.org/abs/2609.36630v1 Summary:This work provides a comprehensive taxonomy…
  • EP525: How AnyAct Fixes Desktop AI Agents · 22 min · Saison 1 Épisode 525Title: AnyAct: Universal Action for Self-Evolving AgentsSource: http://arxiv.org/abs/2609.37025v1 Summary: This paper introduces a foundational framework for universal action…
  • EP524: How SkillRubric Helps AI Grade Itself · 22 min · Saison 1 Épisode 524Title: SkillRubric: Co-Evolving Actor Guidance and Evaluator Rubrics for Multimodal Agents Source: http://arxiv.org/abs/2609.34557v1Summary: This paper introduces a foundational…
  • EP523: How Subgoals Stop AI Belief Drift · 21 min · Saison 1 Épisode 523Title: SGG-ReflAct: Sub-Goal Guided ReflAct with Structured Planning for Reliable Long-Horizon Reasoning Source: http://arxiv.org/abs/2609.34548v1Summary: This work establishes…
  • EP522: Why Limited Memory Makes AI Smarter · 22 min · Saison 1 Épisode 522Title: Continuous Context Management Source: http://arxiv.org/abs/2609.35540v1Summary: This paper introduces a foundational mechanism for dynamically managing context streams in…
  • EP521: HTN Planning Fixes AI Tool Orchestration · 21 min · Saison 1 Épisode 521Title: HTN Planning as a Coordination Layer for Multi-Server MCP Tool Orchestration Source: http://arxiv.org/abs/2609.33731v1 Summary:This work presents Hierarchical Task Network…

En direct, dernière heure : 11 099 titres sur 5 399 stations dans 145 pays

Découvrez le monde par la radio

Plus de façons de découvrir (9)

Vous avez entendu quelque chose de bien ? Enregistrez-le, ou ouvrez-le directement dans Spotify.

El Sofoke RadioContinuer l’écoute