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] From 128K to 4M: Efficient Training of Ultra-Long Context Large Language Models

8 mins; April 08, 2025

From 128K to 4M: Efficient Training of Ultra-Long Context Large Language Models

22 mins; April 08, 2025

[QA] Hogwild! Inference: Parallel LLM Generation via Concurrent Attention

6 mins; April 08, 2025

Hogwild! Inference: Parallel LLM Generation via Concurrent Attention

15 mins; April 08, 2025

[QA] Can ChatGPT Learn My Life From a Week of First-Person Video?

7 mins; April 08, 2025

Can ChatGPT Learn My Life From a Week of First-Person Video?

8 mins; April 08, 2025

[QA] Using Attention Sinks to Identify and Evaluate Dormant Heads in Pretrained LLMs

8 mins; April 08, 2025

Using Attention Sinks to Identify and Evaluate Dormant Heads in Pretrained LLMs

17 mins; April 08, 2025

[QA] Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models

8 mins; April 06, 2025

Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models

25 mins; April 06, 2025

[QA] Agentic Knowledgeable Self-awareness

7 mins; April 06, 2025

Agentic Knowledgeable Self-awareness

18 mins; April 06, 2025

[QA] Inference-Time Scaling for Generalist Reward Modeling

8 mins; April 05, 2025

Inference-Time Scaling for Generalist Reward Modeling

18 mins; April 05, 2025

[QA] Multi-Token Attention

8 mins; April 05, 2025

Multi-Token Attention

18 mins; April 05, 2025

[QA] Visual Jenga: Discovering Object Dependencies via Counterfactual Inpainting

6 mins; March 29, 2025

Visual Jenga: Discovering Object Dependencies via Counterfactual Inpainting

16 mins; March 29, 2025

[QA] Wan: Open and Advanced Large-Scale Video Generative Models

8 mins; March 28, 2025

Wan: Open and Advanced Large-Scale Video Generative Models

64 hours 43 mins; March 28, 2025

[QA] UI-R1: Enhancing Action Prediction of GUI Agents by Reinforcement Learning

8 mins; March 28, 2025

UI-R1: Enhancing Action Prediction of GUI Agents by Reinforcement Learning

16 mins; March 28, 2025

[QA] SWI: Speaking with Intent in Large Language Models

7 mins; March 27, 2025

SWI: Speaking with Intent in Large Language Models

9 mins; March 27, 2025

[QA] Unified Multimodal Discrete Diffusion

7 mins; March 27, 2025

Unified Multimodal Discrete Diffusion

20 mins; March 27, 2025

[QA] Self-Supervised Learning of Motion Concepts by Optimizing Counterfactuals

17 mins; March 26, 2025

Self-Supervised Learning of Motion Concepts by Optimizing Counterfactuals

17 mins; March 26, 2025

[QA] Open Deep Search: Democratizing Search with Open-source Reasoning Agents

7 mins; March 26, 2025

Open Deep Search: Democratizing Search with Open-source Reasoning Agents

17 mins; March 26, 2025

[QA] LookAhead Tuning: Safer Language Models via Partial Answer Previews

7 mins; March 25, 2025

LookAhead Tuning: Safer Language Models via Partial Answer Previews

8 mins; March 25, 2025

[QA] ReSearch: Learning to Reason with Search for LLMs via Reinforcement Learning

7 mins; March 25, 2025

ReSearch: Learning to Reason with Search for LLMs via Reinforcement Learning

13 mins; March 25, 2025

[QA] FFN Fusion: Rethinking Sequential Computation in Large Language Models

8 mins; March 24, 2025

FFN Fusion: Rethinking Sequential Computation in Large Language Models

21 mins; March 24, 2025

[QA] Modifying Large Language Model Post-Training for Diverse Creative Writing

8 mins; March 24, 2025

Modifying Large Language Model Post-Training for Diverse Creative Writing

18 mins; March 24, 2025

[QA] Users Favor LLM-Generated Content—Until They Know It's AI

6 mins; March 23, 2025

Users Favor LLM-Generated Content—Until They Know It's AI

7 mins; March 23, 2025

[QA] Sparse Logit Sampling: Accelerating Knowledge Distillation in LLMs

7 mins; March 23, 2025

Sparse Logit Sampling: Accelerating Knowledge Distillation in LLMs

20 mins; March 23, 2025

[QA] DAPO: An Open-Source LLM Reinforcement Learning System at Scale

8 mins; March 22, 2025

DAPO: An Open-Source LLM Reinforcement Learning System at Scale

14 mins; March 22, 2025

[QA] SynCity: Training-Free Generation of 3D Worlds

7 mins; March 22, 2025

SynCity: Training-Free Generation of 3D Worlds

13 mins; March 22, 2025

[QA] TULIP: Towards Unified Language-Image Pretraining

7 mins; March 21, 2025

TULIP: Towards Unified Language-Image Pretraining

12 mins; March 21, 2025

[QA] Causal Emergence 2.0: Quantifying emergent complexity

7 mins; March 21, 2025

Causal Emergence 2.0: Quantifying emergent complexity

31 mins; March 21, 2025

[QA] Exploring the Hidden Reasoning Process of Large Language Models by Misleading Them

9 mins; March 20, 2025

Exploring the Hidden Reasoning Process of Large Language Models by Misleading Them

22 mins; March 20, 2025

[QA] Time After Time: Deep-Q Effect Estimation for Interventions on When and What to do

7 mins; March 20, 2025

Time After Time: Deep-Q Effect Estimation for Interventions on When and What to do

24 mins; March 20, 2025

[QA] Cube: A Roblox View of 3D Intelligence

7 mins; March 19, 2025

Cube: A Roblox View of 3D Intelligence

15 mins; March 19, 2025

[QA] SWEET-RL: Training Multi-Turn LLM Agents on Collaborative Reasoning Tasks

7 mins; March 19, 2025

SWEET-RL: Training Multi-Turn LLM Agents on Collaborative Reasoning Tasks

20 mins; March 19, 2025

[QA] Measuring AI Ability to Complete Long Tasks

7 mins; March 19, 2025

Measuring AI Ability to Complete Long Tasks

44 mins; March 19, 2025

[QA] Impossible Videos

8 mins; March 19, 2025

Impossible Videos

15 mins; March 19, 2025

[QA] SuperBPE: Space Travel for Language Models

7 mins; March 17, 2025

SuperBPE: Space Travel for Language Models

16 mins; March 17, 2025

[QA] xLSTM 7B: A Recurrent LLM for Fast and Efficient Inference

7 mins; March 17, 2025

xLSTM 7B: A Recurrent LLM for Fast and Efficient Inference

18 mins; March 17, 2025

[QA] PLADIS: Pushing the Limits of Attention in Diffusion Models at Inference Time by Leveraging Sparsity

7 mins; March 17, 2025

PLADIS: Pushing the Limits of Attention in Diffusion Models at Inference Time by Leveraging Sparsity

19 mins; March 17, 2025

[QA] Auditing language models for hidden objectives

8 mins; March 17, 2025

Auditing language models for hidden objectives

37 mins; March 17, 2025

[QA] Alias-Free Latent Diffusion Models: Improving Fractional Shift Equivariance of Diffusion Latent Space

8 mins; March 15, 2025

Alias-Free Latent Diffusion Models: Improving Fractional Shift Equivariance of Diffusion Latent Space

20 mins; March 15, 2025

[QA] New Trends for Modern Machine Translation with Large Reasoning Models

7 mins; March 15, 2025

New Trends for Modern Machine Translation with Large Reasoning Models

15 mins; March 15, 2025

[QA] Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning

8 mins; March 14, 2025

Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning

17 mins; March 14, 2025

[QA] Long Context Tuning for Video Generation

8 mins; March 14, 2025

Long Context Tuning for Video Generation

15 mins; March 14, 2025

[QA] Transformers without Normalization

7 mins; March 14, 2025

Transformers without Normalization

12 mins; March 14, 2025

[QA] Charting and Navigating Hugging Face's Model Atlas

7 mins; March 14, 2025

Charting and Navigating Hugging Face's Model Atlas

13 mins; March 14, 2025

[QA] I Predict Therefore I Am: Is Next Token Prediction Enough to Learn Human-Interpretable Concepts from Data?

8 mins; March 13, 2025

I Predict Therefore I Am: Is Next Token Prediction Enough to Learn Human-Interpretable Concepts from Data?

12 mins; March 13, 2025

[QA] Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models

7 mins; March 13, 2025

Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models

18 mins; March 13, 2025

[QA] Gemini Embedding: Generalizable Embeddings from Gemini

8 mins; March 12, 2025

Gemini Embedding: Generalizable Embeddings from Gemini

16 mins; March 12, 2025

[QA] Inductive Moment Matching

9 mins; March 11, 2025

Inductive Moment Matching

23 mins; March 11, 2025

[QA] Optimizing Test-Time Compute via Meta Reinforcement Fine-Tuning

7 mins; March 11, 2025

Optimizing Test-Time Compute via Meta Reinforcement Fine-Tuning

30 mins; March 11, 2025

[QA] Sketch-of-Thought: Efficient LLM Reasoning with Adaptive Cognitive-Inspired Sketching

7 mins; March 09, 2025

Sketch-of-Thought: Efficient LLM Reasoning with Adaptive Cognitive-Inspired Sketching

19 mins; March 09, 2025

[QA] Continual Pre-training of MoEs: How robust is your router?

8 mins; March 09, 2025

Continual Pre-training of MoEs: How robust is your router?

23 mins; March 09, 2025

[QA] HoT: Highlighted Chain of Thought for Referencing Supporting Facts from Inputs

6 mins; March 08, 2025

HoT: Highlighted Chain of Thought for Referencing Supporting Facts from Inputs

13 mins; March 08, 2025

[QA] Boosting Blockchain Throughput: Parallel EVM Execution with Asynchronous Storage for Reddio

7 mins; March 08, 2025

Boosting Blockchain Throughput: Parallel EVM Execution with Asynchronous Storage for Reddio

26 mins; March 08, 2025

Click here to see more

Arxiv Papers

Useful pages

Find a tutor

Languages