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.

Be sure to visit the podcast's website and support the creator: podcasters.spotify.com

Author

Enoch H. Kang

Category

Technology

Podcast website

podcasters.spotify.com

Latest episode

Nov 19, 2026

Where to listen?

Podcasts in the app Replaio Radio Coming soon

Podcasts are coming to the app soon. Install now and be the first to see a whole new take on podcasts

Get it on Google Play Install for free Android 10M+ downloads · 4.7+ rating iOS soon

Episodes

Multi-Objective Preference Optimization: Improving Human Alignment of Generative Models 22.05.2025

This paper introduces Multi-Objective Preference Optimization (MOPO) , a novel algorithm designed to align large language models with complex human preferences that involve multiple, potentially conflicting goals like helpfulness and harmlessness. Unlike prior methods that often reduce multi-objective alignment to a single score, MOPO frames the problem as a constrained optimization, maximizing a...

End-to-End Learning for Stochastic Optimization: A Bayesian Perspective 21.05.2025

This paper present research on end-to-end learning for stochastic optimization , focusing on a Bayesian perspective . The authors propose that the standard algorithm used in this field has a Bayesian interpretation , effectively training a map that performs a posterior Bayes action . Building on this understanding, they introduce new algorithms for training decision-making tools for problems invol...

TEXTGRAD: Automatic Differentiation via Text 21.05.2025

This collection of excerpts introduces TEXTGRAD , a novel framework that applies the concept of automatic differentiation to complex AI systems composed of multiple large language models and other components. Instead of using numerical gradients like traditional deep learning, TEXTGRAD employs natural language feedback from LLMs to guide the optimization process. The framework, designed with PyTor...

Steering off Course: Reliability Challenges in Steering Language Models 20.05.2025

We investigate the reliability of language model (LM) steering methods , which aim to modify model behavior without retraining. Researchers examined three techniques— DoLa, function vectors, and task vectors —on a wide range of LMs, finding that their effectiveness varies significantly across models and tasks. Contrary to prior research that suggested consistent performance or localization of func...

Past-Token Prediction for Long-Context Robot Policies 20.05.2025

This research presents Past Token Prediction (PTP) , an auxiliary technique designed to improve long-context diffusion policies for robots learning tasks through imitation. The core idea is to explicitly train the policy to predict past actions along with future ones, which helps address the issue of modern diffusion policies failing to capture strong temporal dependencies . A multi-stage training...

Recovering Coherent Event Probabilities from LLM Embeddings 20.05.2025

This paper details research on recovering coherent event probabilities from Large Language Model (LLM) embeddings , as current LLMs often produce probability judgments in text that violate the axioms of probability theory. The authors propose a novel unsupervised learning method using a variational autoencoder (VAE) approach to enforce axiomatic constraints, specifically the additive rule for comp...

Systematic Meta-Abilities Alignment in Large Reasoning Models 20.05.2025

This academic paper proposes a method to improve the reasoning abilities of Large Reasoning Models (LRMs) by moving beyond inconsistent emergent behaviors. The authors introduce a system to explicitly train models in three key meta-abilities : deduction, induction, and abduction, using automatically generated, verifiable tasks. Their three-stage pipeline involves individual alignment of these abil...

Predictability Shapes Adaptation: An Evolutionary Perspective on Modes of Learning in Transformers 20.05.2025

This paper investigates how the predictability of the training environment influences the balance between two distinct learning modes in Transformer models: in-weights learning (IWL) , which is analogous to genetic encoding, and in-context learning (ICL) , which is compared to phenotypic plasticity. Drawing parallels from evolutionary biology, the authors explore how environmental stability (consi...

Efficient Exploration for LLMs 19.05.2025

This Google DeepMind paper investigates efficient exploration strategies for improving large language models (LLMs) through reinforcement learning from human feedback (RLHF) . The authors propose and evaluate various active exploration algorithms, contrasting them with passive methods. Their experiments, using a human preference simulator and the Gemini Nano model, demonstrate that active explorat...

Rankers, Judges, and Assistants: Towards Understanding the Interplay of LLMs in Information Retrieval Evaluation 18.05.2025

This paper, authored by Google DeepMind researchers, explores the increasing reliance on large language models (LLMs) within information retrieval (IR) systems. It examines how LLMs function as rankers, judges, and assistants , powering various aspects from content creation to evaluation. The paper highlights the potential for biases to emerge from the interaction of these LLM-based components , p...

Bayesian Concept Bottlenecks with LLM Priors 17.05.2025

This document introduces **BC-LLM**, a novel method for creating **Concept Bottleneck Models (CBMs)** that are both accurate and interpretable. Traditional CBMs rely on a predefined set of concepts, limiting their effectiveness and requiring significant human effort. **BC-LLM** addresses this by integrating **Large Language Models (LLMs)** within a **Bayesian framework** to iteratively discover re...

Transformers for In-Context Reinforcement Learning 17.05.2025

This paper **explores the theoretical underpinnings of using transformer networks for in-context reinforcement learning (ICRL)**. The authors propose a **general framework for supervised pretraining in meta-RL**, encompassing existing methods like Algorithm Distillation and Decision-Pretrained Transformers. They demonstrate that transformers can **efficiently approximate classical RL algorithms**...

Evaluating Large Language Models Across the Lifecycle 17.05.2025

The sources discuss the importance of robust evaluation for Large Language Models (LLMs) throughout their lifecycle, highlighting a shift from traditional software testing methods due to the non-deterministic nature of LLMs. They cover various evaluation methodologies and metrics , including quantitative, qualitative, and the emerging "LLM-as-a-Judge" approach, while also acknowledging t...

Active Ranking from Human Feedback with DopeWolfe 16.05.2025

This research explores the challenge of learning human preferences over a large set of items using a limited number of ranked comparisons. The authors frame this as learning a Plackett-Luce model from K-way comparisons where K is much smaller than the total number of items. To address the computational complexity of selecting the most informative K-item subsets for comparison, they propose a novel...

Optimal Designs for Preference Elicitation 16.05.2025

This research explores efficient methods for gathering human feedback to build accurate preference models, crucial for modern AI development, especially large language models. Focusing on optimal designs , a technique for determining the most informative data to collect, the paper adapts this approach to lists of items representing potential questions and answers. The methodology is shown to be ef...

Dual Active Learning for Reinforcement Learning from Human Feedback 16.05.2025

This document introduces a novel method for improving the alignment of large language models (LLMs) with human preferences using Reinforcement Learning from Human Feedback (RLHF) . The core contribution is a dual active reward learning algorithm that strategically selects both conversations to be labeled and the most appropriate human teachers to provide that feedback, thereby optimizing the data...

Active Learning for Direct Preference Optimization 16.05.2025

This document explores active learning strategies for Direct Preference Optimization (DPO) , a method for aligning large language models (LLMs) with human preferences by directly optimizing the policy based on feedback. The authors propose a framework and two algorithms, ADPO and ADPO+ , designed for both online collection of new feedback and offline selection from existing feedback, aiming to eff...

Active Preference Optimization for RLHF 16.05.2025

This document introduces Active Preference Optimization (APO) , an algorithm designed to enhance the sample efficiency of Reinforcement Learning from Human Feedback (RLHF) for Large Language Models (LLMs) . The authors highlight the costly bottleneck of collecting high-quality human preference data in current RLHF methods, which often rely on uniform sampling of prompt-generation pairs, leading to...

Test-Time Alignment of Diffusion Models without reward over-optimization 16.05.2025

This text introduces Diffusion Alignment as Sampling (DAS) , a novel approach for aligning diffusion models with desired characteristics by treating the problem as sampling from a reward-aligned distribution. DAS utilizes a Sequential Monte Carlo (SMC) framework enhanced with tempering and a specially designed proposal distribution to efficiently generate high-reward samples without requiring addi...

Test-Time Preference Optimization: On-the-Fly Alignment via Iterative Textual Feedback 16.05.2025

This academic paper introduces Test-Time Preference Optimization (TPO) , a novel method for improving the performance and safety alignment of large language models during inference without altering their core parameters. Unlike traditional alignment techniques that modify the model during training using numerical gradients, TPO leverages the model's own abilities to interpret numerical reward...

GenARM: Reward Guided Generation with Autoregressive Reward Model for Test-time Alignment 16.05.2025

This arXiv paper, titled  GenARM: Reward Guided Generation with Autoregressive Reward Model for Test-time Alignment , introduces a  novel approach  for aligning  Large Language Models (LLMs)  with human preferences  during the inference stage , without requiring expensive retraining. The authors propose the  Autoregressive Reward Model  to address the limitations of existing methods that use  traj...

Advantage-Weighted Regression: Simple and Scalable Off-Policy RL 16.05.2025

This paper introduces and explains Advantage-Weighted Regression (AWR) , a simple and scalable off-policy reinforcement learning algorithm that utilizes standard supervised learning techniques. The paper details AWR's theoretical basis , highlighting its connection to constrained policy optimization and its ability to effectively handle off-policy data through experience replay . The authors d...

Can RLHF be More Efficient with Imperfect Reward Models? A Policy Coverage Perspective 16.05.2025

This research explores ways to make Reinforcement Learning from Human Feedback (RLHF) more sample-efficient by leveraging imperfect reward models . The authors identify a key property of the KL-regularized RLHF objective , showing that a policy's ability to cover the optimal policy is linked to its sub-optimality, which suggests that higher policy value indicates better coverage . Building on...

Transformers can be used for in-context linear regression in the presence of endogeneity 15.05.2025

This paper explores how  transformers  can be used for  in-context linear regression  in the presence of  endogeneity . The authors demonstrate theoretically that transformers can effectively handle endogeneity by implementing  instrumental variables (IV)  techniques, specifically the  two-stage least squares (2SLS)  method, through a gradient-based approach that converges exponentially. They prop...

Bayesian Concept Bottlenecks with LLM Priors 15.05.2025

This document introduces BC-LLM , a novel method for creating Concept Bottleneck Models (CBMs) that are both accurate and interpretable. Traditional CBMs rely on a predefined set of concepts, limiting their effectiveness and requiring significant human effort. BC-LLM addresses this by integrating Large Language Models (LLMs) within a Bayesian framework to iteratively discover relevant concepts. Th...

Listen to the Best AI papers explained podcast in Replaio

Radio and podcasts in one app - free, with no sign-up. Install today and do not miss the launch

Get it on Google Play

Replaio is not a podcast publisher; show names, artwork and audio belong to their authors and are distributed through public RSS feeds.