1. Podcasts
  2. Technology
  3. Best AI papers explained
Best AI papers explained cover art

530 episodes

Best AI papers explained, page 22

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 22

Human-AI Matching: The Limits of Algorithmic Search

AI “clones” trained on an individual’s digital traces could be used to match people for romance, jobs, housing, childcare, tutors, or therapists; the episode explains a research paper’s limits on…

25 Jun 2025 · 15m · 6 chapters

Uncertainty Quantification Needs Reassessment for Large-language Model Agents

The episode argues that uncertainty quantification for large-language-model agents needs reassessment because hallucinations are inherent to generative models, and the traditional aleatoric vs…

25 Jun 2025 · 19m · 11 chapters

Bayesian Meta-Reasoning for Robust LLM Generalization

The episode argues current LLMs are strong pattern matchers but lack meta-reasoning—thinking about thinking—so they hallucinate confidently, fail under small prompt changes, and generalize poorly.

25 Jun 2025 · 20m · 9 chapters

General Intelligence Requires Reward-based Pretraining

The episode discusses a research paper arguing that today’s large language models have fragile, non-generalizable reasoning because supervised pretraining entangles “knowledge” with “how to reason.”…

25 Jun 2025 · 17m · 12 chapters

Deep Learning is Not So Mysterious or Different

Explains major “mysteries” of deep learning generalization—benign overfitting, over-parameterization, and double descent—using unifying ideas like simplicity/compression bias and “soft inductive…

25 Jun 2025 · 22m · 13 chapters

AI Agents Need Authenticated Delegation

The episode explains “authenticated delegation” for AI agents—how to prove an agent’s identity, that it acts on behalf of a specific user, and that it has only task-specific permissions (not carte…

25 Jun 2025 · 19m · 10 chapters

Probabilistic Modelling is Sufficient for Causal Inference

Argues that causal inference (interventions and counterfactuals) can be fully handled within standard probabilistic modeling, using the principle “write down the probability of everything” (David…

25 Jun 2025 · 22m · 8 chapters

Not All Explanations for Deep Learning Phenomena Are Equally Valuable

A discussion of a position paper arguing that “deep learning phenomena” (edge cases like double descent, grokking, and lottery ticket hypothesis) should be studied for their ability to refine broad,…

25 Jun 2025 · 19m · 10 chapters

e3: Learning to Explore Enables Extrapolation of Test-Time Compute for LLMs

E3 (“Exploration Enables Extrapolation”) teaches LLMs to use more test-time compute by learning in-context exploration—chaining generation, verification, and refinement to improve beyond the training…

17 Jun 2025 · 14m · 8 chapters

Extrapolation by Association: Length Generalization Transfer in Transformers

“Extrapolation by association” for length generalization in transformers—how co-training on a related auxiliary task with longer inputs lets a model extrapolate to longer lengths on a main task it…

17 Jun 2025 · 12m · 6 chapters

Uncovering Causal Hierarchies in Language Model Capabilities

The episode explains research on evaluating LLM capabilities beyond single benchmark scores, using PCA/HCA on thousands of models from the Open LLM leaderboard to infer a low-rank “causal hierarchy”…

17 Jun 2025 · 19m · 9 chapters

Generalization or Hallucination? Understanding Out-of-Context Reasoning in Transformers

Generalization vs hallucination in transformers, arguing both come from one mechanism called out-of-context reasoning (OCR): learning implications via concept co-occurrence even when no…

17 Jun 2025 · 14m · 5 chapters

Improving Treatment Effect Estimation with LLM-Based Data Augmentation

The episode explains GATE, a framework for improving conditional average treatment effect (CATE) estimation under data scarcity and covariate shift.

17 Jun 2025 · 15m · 9 chapters

LLM Numerical Prediction Without Auto-Regression

The episode discusses research showing that LLMs (example: LLaMA-28B) encode numerical predictions and uncertainty in internal hidden states before autoregressive token generation, enabling faster,…

17 Jun 2025 · 15m · 11 chapters

Why in-context learning models are good few-shot learners?

Explains why transformer-based in-context learning (ICL) enables strong few-shot learning without updating model weights, framing it as meta-learning.

17 Jun 2025 · 21m · 10 chapters

Take Caution in Using LLMs as Human Surrogates: Scylla Ex Machina∗

The episode critiques using large language models (LLMs) as “human surrogates” for social science research, focusing on a paper titled “Take Caution in LLMs as Human Surrogates: Scylla Ex Machina.”…

14 Jun 2025 · 28m · 14 chapters

The Logic of Machines: The AI Reasoning Debate

Whether today’s “AI reasoning” is genuine multi-step cognitive reasoning (deduction, planning, understanding) or sophisticated pattern matching that only mimics logic, with implications for trust,…

12 Jun 2025 · 31m · 14 chapters

Layer by Layer: Uncovering Hidden Representations in Language Models

The episode explains a research finding that large language models often encode their most useful representations in intermediate (middle) layers rather than the final layer.

12 Jun 2025 · 13m · 5 chapters

Causal Attribution Analysis for Continuous Outcomes

The episode explains Luo, Yu, Lu, Shi, and Jing’s “Causal Attribution Analysis for Continuous Outcomes,” which answers a retrospective “why” question for continuous results (e.g., blood pressure,…

12 Jun 2025 · 18m · 7 chapters

Training a Generally Curious Agent

The episode discusses the paper “Training a Generally Curious Agent” (Fahim Tajwar et al., aimed at ICML 2025), arguing current LLMs are strong pattern matchers but weak at strategic exploration in…

12 Jun 2025 · 14m · 8 chapters

Estimation of Treatment Effects Under Nonstationarity via Truncated Difference-in-Q’s

The episode explains why standard A/B testing (difference in means) can be badly biased in non-stationary, interactive systems where actions cause temporal interference and the environment changes…

12 Jun 2025 · 21m · 9 chapters

Strategy Coopetition Explains the Emergence and Transience of In-Context Learning

Explains why in-context learning (ICL) in transformer models can appear during training but later fade, using a “strategy coopetition” framework.

12 Jun 2025 · 19m · 14 chapters

Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs

The episode discusses a research paper claiming that narrow fine-tuning of already-aligned LLMs on a single harmful behavior can cause “emergent misalignment,” where the model becomes broadly…

11 Jun 2025 · 17m · 7 chapters

Agentic Supernet for Multi-agent Architecture Search

The episode explains MAAS (Multi-Agent Architecture Search via Agentic SuperNet), a framework for automatically designing multi-agent LLM workflows that adapt per query instead of using one fixed…

11 Jun 2025 · 18m · 12 chapters