LanguaTalk

Arxiv Papers

Running out of time to catch up with new arXiv papers? We take the most impactful papers and present them as convenient podcasts. If you're a visual learner, we offer these papers in an engaging video format. Our service fills the gap between overly brief paper summaries and time-consuming full paper reads. You gain academic insights in a time-efficient, digestible format. Code behind this work: https://github.com/imelnyk/ArxivPapers

[QA] Grokking at the Edge of Numerical Stability

7 mins; January 08, 2025

Grokking at the Edge of Numerical Stability

16 mins; January 08, 2025

[QA] ComMer: a Framework for Compressing and Merging User Data for Personalization

7 mins; January 07, 2025

ComMer: a Framework for Compressing and Merging User Data for Personalization

16 mins; January 07, 2025

[QA] Entropy-Guided Attention for Private LLMs

7 mins; January 07, 2025

Entropy-Guided Attention for Private LLMs

13 mins; January 07, 2025

[QA] Easing Optimization Paths: a Circuit Perspective

7 mins; January 06, 2025

Easing Optimization Paths: a Circuit Perspective

9 mins; January 06, 2025

[QA] Randomly Sampled Language Reasoning Problems Reveal Limits of LLMs

7 mins; January 06, 2025

Randomly Sampled Language Reasoning Problems Reveal Limits of LLMs

16 mins; January 06, 2025

[QA] Enhancing Reasoning through Process Supervision with Monte Carlo Tree Search

6 mins; January 05, 2025

Enhancing Reasoning through Process Supervision with Monte Carlo Tree Search

7 mins; January 05, 2025

[QA] Predicting the Performance of Black-box LLMs through Self-Queries

7 mins; January 05, 2025

Predicting the Performance of Black-box LLMs through Self-Queries

20 mins; January 05, 2025

[QA] On Unifying Video Generation and Camera Pose Estimation

8 mins; January 03, 2025

On Unifying Video Generation and Camera Pose Estimation

21 mins; January 03, 2025

[QA] An analytic theory of creativity in convolutional diffusion models

8 mins; January 03, 2025

An analytic theory of creativity in convolutional diffusion models

22 mins; January 03, 2025

[QA] Finding Missed Code Size Optimizations in Compilers using LLMs

7 mins; January 03, 2025

Finding Missed Code Size Optimizations in Compilers using LLMs

18 mins; January 03, 2025

[QA] Titans: Learning to Memorize at Test Time

7 mins; January 03, 2025

Titans: Learning to Memorize at Test Time

30 mins; January 03, 2025

[QA] Adaptive Batch Size Schedules for Distributed Training of Language Models with Data and Model Parallelism

7 mins; December 30, 2024

Adaptive Batch Size Schedules for Distributed Training of Language Models with Data and Model Parallelism

20 mins; December 30, 2024

[QA] Functional Risk Minimization

8 mins; December 30, 2024

Functional Risk Minimization

25 mins; December 30, 2024

[QA] HuatuoGPT-o1, Towards Medical Complex Reasoning with LLMs

8 mins; December 30, 2024

HuatuoGPT-o1, Towards Medical Complex Reasoning with LLMs

17 mins; December 30, 2024

InfAlign: Inference-aware language model alignment

27 mins; December 30, 2024

[QA] Consistency Checks for Language Model Forecasters

7 mins; December 25, 2024

Consistency Checks for Language Model Forecasters

18 mins; December 25, 2024

[QA] Deliberation in Latent Space via Differentiable Cache Augmentation

7 mins; December 24, 2024

Deliberation in Latent Space via Differentiable Cache Augmentation

20 mins; December 24, 2024

[QA] Automating the Search for Artificial Life with Foundation Models

7 mins; December 24, 2024

Automating the Search for Artificial Life with Foundation Models

20 mins; December 24, 2024

[QA] LLMs for Literature Review: Are we there yet?

6 mins; December 22, 2024

LLMs for Literature Review: Are we there yet?

5 mins; December 22, 2024

[QA] WebLLM: A High-Performance In-Browser LLM Inference Engine

6 mins; December 22, 2024

WebLLM: A High-Performance In-Browser LLM Inference Engine

7 mins; December 22, 2024

[QA] A Survey on LLM Inference-Time Self-Improvement

7 mins; December 21, 2024

A Survey on LLM Inference-Time Self-Improvement

23 mins; December 21, 2024

[QA] Tokenisation is NP-Complete

7 mins; December 21, 2024

Tokenisation is NP-Complete

17 mins; December 21, 2024

[QA] The Open-Source Advantage in Large Language Models (LLMs)

8 mins; December 21, 2024

The Open-Source Advantage in Large Language Models (LLMs)

13 mins; December 21, 2024

[QA] Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference

7 mins; December 21, 2024

Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference

20 mins; December 21, 2024

[QA] DriveGPT: Scaling Autoregressive Behavior Models for Driving

8 mins; December 20, 2024

DriveGPT: Scaling Autoregressive Behavior Models for Driving

19 mins; December 20, 2024

[QA] MetaMorph: Multimodal Understanding and Generation via Instruction Tuning

8 mins; December 20, 2024

MetaMorph: Multimodal Understanding and Generation via Instruction Tuning

16 mins; December 20, 2024

[QA] Byte Latent Transformer: Patches Scale Better Than Tokens

7 mins; December 17, 2024

Byte Latent Transformer: Patches Scale Better Than Tokens

40 mins; December 17, 2024

[QA] Transformers Struggle to Learn to Search

7 mins; December 08, 2024

Transformers Struggle to Learn to Search

19 mins; December 08, 2024

[QA] Navigation World Models

7 mins; December 07, 2024

Navigation World Models

19 mins; December 07, 2024

[QA] Motion Prompting: Controlling Video Generation with Motion Trajectories

7 mins; December 07, 2024

Motion Prompting: Controlling Video Generation with Motion Trajectories

16 mins; December 07, 2024

[QA] Infinity: Scaling Bitwise AutoRegressive Modeling for High-Resolution Image Synthesis

8 mins; December 06, 2024

Infinity: Scaling Bitwise AutoRegressive Modeling for High-Resolution Image Synthesis

21 mins; December 06, 2024

[QA] NVILA: Efficient Frontier Visual Language Models

7 mins; December 06, 2024

NVILA: Efficient Frontier Visual Language Models

23 mins; December 06, 2024

[QA] o1-Coder: an o1 Replication for Coding

9 mins; December 03, 2024

o1-Coder: an o1 Replication for Coding

19 mins; December 03, 2024

[QA] Efficient Track Anything

7 mins; December 03, 2024

Efficient Track Anything

19 mins; December 03, 2024

[QA] Reverse Thinking Makes LLMs Stronger Reasoners

8 mins; December 01, 2024

Reverse Thinking Makes LLMs Stronger Reasoners

16 mins; December 01, 2024

[QA] Critical Tokens Matter: Token-Level Contrastive Estimation Enhence LLM’s Reasoning Capability

7 mins; December 01, 2024

Critical Tokens Matter: Token-Level Contrastive Estimation Enhence LLM’s Reasoning Capability

12 mins; December 01, 2024

[QA] JetFormer: an autoregressive generative model of raw images and text

7 mins; December 01, 2024

JetFormer: an autoregressive generative model of raw images and text

23 mins; December 01, 2024

[QA] CAT4D: Create Anything in 4D with Multi-View Video Diffusion Models

7 mins; November 29, 2024

CAT4D: Create Anything in 4D with Multi-View Video Diffusion Models

15 mins; November 29, 2024

Attamba: Attending To Multi-Token States

15 mins; November 28, 2024

[QA] Star Attention: Efficient LLM Inference over Long Sequences

7 mins; November 28, 2024

Star Attention: Efficient LLM Inference over Long Sequences

15 mins; November 28, 2024

[QA] ROICtrl: Boosting Instance Control for Visual Generation

8 mins; November 28, 2024

ROICtrl: Boosting Instance Control for Visual Generation

20 mins; November 28, 2024

[QA] Solaris: A Foundation Model of the Sun

7 mins; November 26, 2024

Solaris: A Foundation Model of the Sun

16 mins; November 26, 2024

[QA] Even Sparser Graph Transformers

7 mins; November 26, 2024

Even Sparser Graph Transformers

21 mins; November 26, 2024

[QA] Understanding LLM Embeddings for Regression

6 mins; November 24, 2024

Understanding LLM Embeddings for Regression

12 mins; November 24, 2024

[QA] Loss-to-Loss Prediction: Scaling Laws for All Datasets

7 mins; November 24, 2024

Loss-to-Loss Prediction: Scaling Laws for All Datasets

21 mins; November 24, 2024

[QA] When Precision Meets Position: BFloat16 Breaks Down RoPE in Long-Context Training

7 mins; November 22, 2024

When Precision Meets Position: BFloat16 Breaks Down RoPE in Long-Context Training

21 mins; November 22, 2024

[QA] SAMURAI: Adapting Segment Anything Model for Zero-Shot Visual Tracking with Motion-Aware Memory

7 mins; November 22, 2024

SAMURAI: Adapting Segment Anything Model for Zero-Shot Visual Tracking with Motion-Aware Memory

18 mins; November 22, 2024

[QA] Do I Know This Entity? Knowledge Awareness and Hallucinations in Language Models

7 mins; November 22, 2024

Do I Know This Entity? Knowledge Awareness and Hallucinations in Language Models

13 mins; November 22, 2024

[QA] Hymba: A Hybrid-head Architecture for Small Language Models

8 mins; November 22, 2024

Hymba: A Hybrid-head Architecture for Small Language Models

21 mins; November 22, 2024

[QA] Is Your LLM Secretly a World Model of the Internet? Model-Based Planning for Web Agents

7 mins; November 21, 2024

Is Your LLM Secretly a World Model of the Internet? Model-Based Planning for Web Agents

14 mins; November 21, 2024

[QA] Procedural Knowledge in Pretraining Drives Reasoning in Large Language Models

7 mins; November 20, 2024

Procedural Knowledge in Pretraining Drives Reasoning in Large Language Models

23 mins; November 20, 2024

Click here to see more

Arxiv Papers

Useful pages

Find a tutor

Languages