Podcast · 475 episodes
Best AI papers explained, page 20
by Enoch H. Kang · English
Cut through the noise. We curate and break down the most important AI papers so you don’t have to.
All episodes, page 20
PrefillOnly: An Inference Engine for Prefill-only Workloads in Large Language Model Applications“PrefillOnly” inference engine for LLM applications where each request outputs exactly one token (e.g., yes/no, option A/B/C, label).14 Jul 2025 · 14 min · 4 chapters
A Collectivist, Economic Perspective on AIThe episode argues that today’s AI debate is “noisy” and too individualistic (thinking machines/AGI).14 Jul 2025 · 21 min · 11 chapters
Textual Bayes: Quantifying Uncertainty in LLM-Based SystemsThe episode discusses the paper “Textual Bayes: Quantifying Uncertainty in LLM-Based Systems,” which aims to make LLM outputs more reliable by producing calibrated uncertainty estimates.12 Jul 2025 · 9 min · 6 chapters
The Winner's Curse in Data-Driven DecisionsThe episode explains the “winner’s curse” in inference-then-optimization systems: even with unbiased estimates for each option, selecting the best-performing option based on noisy estimates makes the…11 Jul 2025 · 30 min · 14 chapters
SPIRAL: Self-Play for Reasoning Through Zero-Sum GamesSpiral is a self-play training framework for language-model reasoning in two-player zero-sum, multi-turn “language games,” aiming to reduce reliance on human-labeled data and expert supervision.11 Jul 2025 · 17 min · 9 chapters
Beyond Statistical Learning: Exact Learning Is Essential for General IntelligenceThe episode argues that today’s frontier LLMs (trained via statistical learning/next-token prediction) are misaligned with the deductive, zero-error reasoning needed for general intelligence and…11 Jul 2025 · 22 min · 11 chapters
Aligning Learning and Endogenous Decision-MakingThe episode explains a research paper, “Aligning Learning and Endogenous Decision Making,” arguing that standard AI “predict-then-optimize” fails when actions change the data you observe (endogenous…11 Jul 2025 · 16 min · 11 chapters
Reliable Statistical Inference with Synthetic Data from Large Language ModelsThe episode argues that LLMs can generate fully synthetic survey/text data to overcome the high cost and slow timelines of collecting human social-science data, but only if synthetic data is combined…11 Jul 2025 · 14 min · 12 chapters
Multi-Turn Reinforcement Learning from Human Preference FeedbackMulti-turn reinforcement learning from human preference feedback (RLHF), arguing that training LLMs with feedback on whole conversations beats single-turn feedback because good actions can look bad…10 Jul 2025 · 17 min · 11 chapters
Provably Learning from Language FeedbackHow large language models can learn from detailed natural-language critiques (LLF) instead of scalar reward scores, and when this yields provably faster learning.9 Jul 2025 · 17 min · 8 chapters
Markets with Heterogeneous Agents: Dynamics and Survival of Bayesian vs. No-Regret LearnersHow competing learning algorithms in markets determine “survival” (nonzero wealth share) vs “vanishing” (wealth share goes to zero), comparing Bayesian learners to no-regret learners, and proposing a…5 Jul 2025 · 21 min · 10 chapters
Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric FoundationA July 3, 2025 theoretical framework explains how gradient-based training of neural networks can yield discrete symbolic structures (neurosymbolic reasoning) from continuous optimization.5 Jul 2025 · 14 min · 4 chapters
Causal Abstraction with Lossy RepresentationsProjected abstractions for causal reasoning with lossy representations, addressing the abstract invariance condition (AIC) problem and enabling cross-layer inference (L1 observations to L2…4 Jul 2025 · 26 min · 13 chapters
The Winner's Curse in Data-Driven DecisionsWinner’s curse in data-driven decisions: when teams do “inference, then optimize” (estimate values from data, then pick the best), selection bias from noise makes the chosen option’s estimated lift…4 Jul 2025 · 23 min · 14 chapters
Embodied AI Agents: Modeling the WorldEmbodied AI agents that perceive (vision/audio/touch), model the physical and human “mental” context, plan and act with autonomy, and learn continuously; the episode covers research status (a July 4,…4 Jul 2025 · 29 min · 15 chapters
Beyond Statistical Learning: Exact Learning Is Essential for General IntelligenceThe episode argues that today’s LLMs, trained via statistical learning (next-token prediction), are fundamentally misaligned with the kind of exact, deductive reasoning needed for general…4 Jul 2025 · 20 min · 12 chapters
What Has a Foundation Model Found? Inductive Bias Reveals World ModelsAn “inductive bias probe” tests whether foundation models learn true underlying world rules (Newton) or only learn shortcuts that predict outcomes (Kepler).4 Jul 2025 · 12 min · 6 chapters
Language Bottleneck Models: A Framework for Interpretable Knowledge Tracing and BeyondLanguage Bottleneck Models (LBMs) for knowledge tracing: converting a student’s interaction history into a concise natural-language “knowledge state” summary that preserves predictive power while…3 Jul 2025 · 23 min · 12 chapters
Learning to Explore: An In-Context Learning Approach for Pure ExplorationThe episode explains “pure exploration” in AI and a new method called In-Context Pure Exploration (ICPE), which learns how to choose the most informative next actions without being given explicit…3 Jul 2025 · 17 min · 9 chapters