Enoch H. Kang
Best AI papers explained
Cut through the noise. We curate and break down the most important AI papers so you don’t have to.
Vizitează neapărat site-ul podcastului și susține-i creatorul: podcasters.spotify.com
Autor
Enoch H. Kang
Categorie
Site-ul podcastului
Cel mai nou episod
19 nov. 2026
Unde asculți?
Podcasturi în aplicație Replaio Radio În curândPodcasturile ajung în curând în aplicație. Instaleaz-o acum și fii primul care descoperă o abordare complet nouă a podcasturilor
Episoade
Adaptation of Agentic AI 23.12.2025 13:20
This paper introduces a systematic framework for **agentic AI adaptation**, categorizing research into four distinct paradigms based on whether the **agent** or its **tools** are being optimized. **Agent adaptation** involves updating core models using either **tool-execution signals** for causal feedback or **agent-output signals** for holistic task performance. In contrast, **tool adaptation** f...
Posterior Behavioral Cloning: Pretraining BC Policies for Efficient RL Finetuning 22.12.2025 10:30
This research introduces Posterior Behavioral Cloning (POSTBC), a novel pretraining method designed to enhance reinforcement learning (RL) finetuning for robotic policies. Standard behavioral cloning often fails because it overfits to specific demonstration data, leading to an action coverage deficit that prevents the model from exploring effectively during later stages. To solve this, the authors...
Let’s (not) just put things in Context: Test-Time Training for Long-Context LLMs 21.12.2025 13:45
Large language models often struggle with long-context tasks because the attention mechanism suffers from **score dilution**, where relevant information is overwhelmed by surrounding "distractor" tokens. Researchers found that common **inference-time scaling strategies**, such as generating additional "thinking tokens," fail to solve this problem as context length increases. To address this, the a...
TabPFN-2.5: Advancing the State of the Art in Tabular Foundation Models 20.12.2025 14:30
This paper discusses TabPFN-2.5, a sophisticated tabular foundation model designed to handle diverse datasets with up to 50,000 samples and 2,000 features. This next-generation AI significantly outperforms traditional tree-based models and complex ensembles like AutoGluon in a fraction of the time. The researchers highlight its state-of-the-art performance across various industries, particularly i...
What’s In My Human Feedback? Learning Interpretable Descriptions of Preference Data 19.12.2025 16:14
This paper introduces a method for automatically decoding hidden preferences from language model training data. By utilizing sparse autoencoders, the method translates complex text embeddings into a small set of interpretable features that explain why human annotators prefer one response over another. The research reveals that feedback datasets often contain conflicting signals, such as Reddit use...
Bolmo: Byteifying the Next Generation of Language Models 19.12.2025 13:13
We discuss Bolmo, a groundbreaking family of byte-level language models by AI2 that offers a practical alternative to traditional subword-based tokenization. Developed by the Allen Institute for AI and collaborating universities, these models achieve state-of-the-art performance by "byteifying" existing subword models like OLMo. This innovative process uses a specialized two-stage distillation pro...
What happened with sparse autoencoders? 17.12.2025 30:09
We cover Neel Nanda (Google DeepMind)'s discussion on efficacy and limitations of Sparse Autoencoders (SAEs) as a tool for unsupervised discovery and interpretability in large language models. Initially considered a major breakthrough for breaking down model activations into interpretable, linear concepts, the conversation explores the subsequent challenges and pathologies observed in SAEs, su...
What Matters Right Now in Mechanistic Interpretability 16.12.2025 32:30
We discuss Neel Nanda (Google DeepMind)'s perspectives on the current state and future directions of mechanistic interpretability (MI) in AI research. Nanda discusses major shifts in the field over the past two years, highlighting the improved capabilities and "scarier" nature of modern models, alongside the increasing use of inference time compute and reinforcement learning. A key t...
CLaRa: Bridging Retrieval and Generation with Continuous Latent Reasoning 16.12.2025 14:45
This paper discusses how Retrieval-Augmented Generation (RAG) framework can be designed to overcome the structural issues of separate retrieval and generation modules. The proposed framework, CLaRa, achieves this by employing a **shared latent space** where documents are compressed into concise, continuous memory-token representations, addressing the architectural mismatch and efficiency problems...
Self-Improving AI and Human Co-Improvement for Safer Co-Superintelligence 16.12.2025 13:13
This paper studies "co-improvement" as a safer and faster alternative to the current focus on "autonomous self-improving AI" for achieving superintelligence. This paper argues that instead of AI systems improving themselves without human intervention, the focus should be on building AI that actively collaborates with human researchers across all stages of the research pipeline, from ideation to ev...
Towards a Science of Scaling Agent Systems / Google Deepmind 15.12.2025 15:46
This academic paper by Google Research, Google DeepMind, and the Massachusetts Institute of Technology, systematically evaluates the principles for scaling language model-based agent systems, moving beyond anecdotal evidence that "more agents is all you need." The authors present a controlled evaluation across four diverse agentic benchmarks, testing five canonical architectures—Single-A...
Emergent hierarchical reasoning in LLMs through reinforcement learning 14.12.2025 13:07
This paper discusses how a successful RL fine-tuning uncovers an emergent two-phase hierarchical reasoning dynamic in LLMs, mirroring human cognition by separating high-level strategic planning from low-level procedural execution. The authors argue that conventional RL methods, which apply optimization pressure agnostically to all tokens, are inefficient because they fail to concentrate learning e...
AI revolution finally comes to Relational foundational models for structured data 13.12.2025 14:39
We discuss an interview with Jure Lescovec, co-founder of kumu.ai and a computer science professor at Stanford, regarding the application of foundation models to structured enterprise data. Lescovec explains that traditional **machine learning** methods for this type of data are manual, expensive, and time-consuming, contrasting them with new relational foundation models that leverage a **graph-ba...
REFRAG: Rethinking RAG based Decoding 13.12.2025 13:48
This paperq introduces REFRAG, an innovative and efficient decoding framework specifically designed to accelerate *lRetrieval-Augmented Generation (RAG) in Large Language Models (LLMs) by addressing high latency and memory demands associated with long-context inputs. The core mechanism involves compressing context by representing chunks of retrieved text as single embeddings, significantly shorten...
Provable Long-Range Benefits of Next-Token Prediction 12.12.2025 12:03
This academic paper rigorously investigates the power of next-token prediction for training large language models (LLMs), specifically focusing on Recurrent Neural Networks (RNNs). The core finding is that simply minimizing the next-token log loss during training is sufficient to yield an LLM whose output is computationally indistinguishable from the true training distribution over long sequences...
Jeff Dean on TPUs, AI Research, and Funding 12.12.2025 38:17
We summarize a recent interview with Jeff Dean, a legendary Chief Scientist at Google who has been leading Gemini, focusing on the **evolution and current state of Google's Tensor Processing Units (TPUs)**, including the recent seventh-generation announcement. Dean explains that the initial motivation for TPUs was Google's internal need to handle the massive compute requirements of scaling AI mode...
Latent Debate: surrogate framework for Interpreting LLM Thinking 11.12.2025 15:15
This paper introduces Latent Debate, a novel framework designed to interpret the internal "thinking" processes and address hallucinations in Large Language Models (LLMs). Unlike external methods that rely on multiple models debating, Latent Debate uses implicit internal arguments—supporting and attacking signals—arising within a single model during a single inference. This framework util...
Distribution-calibrated inference time compute for thinking llm-as-a-judge 11.12.2025 11:48
This paper discusses the Distribution-Calibrated Aggregation scheme designed to improve the reliability of "Thinking-LLM-as-a-Judge" systems, which are often used for evaluating generative AI outputs. The core problem addressed is that simply aggregating multiple, noisy individual judgments (e.g., via majority vote) is suboptimal, especially when the judge is allowed to declare a tie. Th...
Principled RL for diffusion LLMs emerges from sequence level perspective 11.12.2025 11:47
This paper establishes sequence-level optimization as the superior paradigm for fine-tuning diffusion LLMs. They introduce a new machine learning framework called ELBO-based Sequence-level Policy Optimization (ESPO), designed to address the fundamental mismatch when applying Reinforcement Learning (RL) to non-autoregressive diffusion Large Language Models (dLLMs). Traditional RL methods rely on to...
Algorithmic Thinking Theory 10.12.2025 17:15
This paper introduce a theoretical framework for studying "algorithmic thinking" in Large Language Models (LLMs), focusing on how iterative refinement and the aggregation of multiple solutions improve performance on complex reasoning tasks, like advanced mathematics problems. This framework formalizes the LLM as a **"reasoning oracle"** that generates new solutions based on a context of previous a...
On the Interplay of Pre-Training, Mid-Training, and RL on Reasoning Language Models 10.12.2025 13:48
This paper details a controlled experimental framework used to examine the interaction between pre-training, mid-training, and reinforcement learning (RL) on the reasoning abilities of language models (LMs). Researchers from Carnegie Mellon University and the Language Technologies Institute utilized a synthetic dataset with explicitly defined reasoning complexity and contextual templates to isolat...
Natural language actor-critic: Scalable off-policy learning in language space 09.12.2025 13:49
This paper introduces Natural Language Actor-Critic (NLAC), a novel off-policy reinforcement learning algorithm designed to train Large Language Model (LLM) agents for complex, multi-turn tasks. NLAC addresses the limitations of traditional methods, which rely on sparse scalar rewards and unstable on-policy training, by employing a generative LLM critic that outputs training signals as natural lan...
Beyond the Transformer: Titans, MIRAS, and the Future of Infinite Context 07.12.2025 38:47
We explore Google's Titans and the MIRAS framework, a new paradigm in sequence modeling that replaces static context compression with active test-time learning. We discuss how Titans utilize deep neural memory modules to update parameters on the fly using a gradient-based "surprise metric," prioritizing unexpected information for long-term storage. We cover the theoretical MIRAS blue...
On the Limits of Test-Time Compute: Sequential Reward Filtering for Better Inference 07.12.2025 13:45
This paper analyzes the fundalmental limitations of Best-of-N (BoN) sampling, proving theoretically that they are suboptimal under a mixture-of-reference-policies model. They propose RF-SeqBoN as a sequential approach that improves efficiency by selectively incorporating only **high-reward generations** back into the LLM's context, thereby concentrating computation on superior policy candidates. B...
The Universal Weight Subspace Hypothesis 07.12.2025 15:54
This paper presents a large-scale empirical analysis supporting **The Universal Weight Subspace Hypothesis**, which posits that deep neural networks, regardless of initialization, task, or domain, converge to remarkably similar low-dimensional parametric subspaces. This research demonstrates that a **small number of principal directions** consistently capture the majority of variance in the weight...
Podcasturi asemănătoare
Replaio nu este editorul podcasturilor; numele emisiunilor, coperțile și materialul audio aparțin autorilor lor și sunt distribuite prin fluxuri RSS publice