Best Retriever Training Podcasts (2024)

1
Episode 185: Jacob Guggenheim - Augmenting Grossing With AI: The Sentinel From Vistapath 36:56

3h ago36:56

36:56

Jacob Guggenheim is head of engineering at Vistapath. Their innovative Sentinel is bringing AI to specimen grossing. Jacob's background in bioengineering and robotics led him to join Vistapath Vistapath's mission is to improve the efficiency and accuracy of the histology process, inspired by a real-world incident The Sentinel is designed to augment…

1
[QA] CoD, Towards an Interpretable Medical Agent using Chain of Diagnosis 8:19

8m ago8:19

8:19

This study presents Chain-of-Diagnosis (CoD) to enhance interpretability in LLM-based medical diagnostics, enabling transparent reasoning and improved performance in diagnosing 9,604 diseases with DiagnosisGPT. https://arxiv.org/abs//2407.13301 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts…

1
CoD, Towards an Interpretable Medical Agent using Chain of Diagnosis 23:15

9m ago23:15

23:15

This study presents Chain-of-Diagnosis (CoD) to enhance interpretability in LLM-based medical diagnostics, enabling transparent reasoning and improved performance in diagnosing 9,604 diseases with DiagnosisGPT. https://arxiv.org/abs//2407.13301 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts…

1
[QA] Recursive Introspection: Teaching Language Model Agents How to Self-Improve 7:44

12m ago7:44

7:44

https://arxiv.org/abs//2407.18219 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016 Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers --- Support this podcast: https://podcasters.spotify.com/pod/show/arxiv-papers/supp…

1
Recursive Introspection: Teaching Language Model Agents How to Self-Improve 39:01

13m ago39:01

39:01

https://arxiv.org/abs//2407.18219 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016 Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers --- Support this podcast: https://podcasters.spotify.com/pod/show/arxiv-papers/supp…

1
[QA] Course-Correction: Safety Alignment Using Synthetic Preferences 7:24

8m ago7:24

7:24

This paper evaluates and enhances large language models' ability to autonomously correct harmful content, introducing the C-EVAL benchmark and a synthetic dataset for effective preference learning. https://arxiv.org/abs//2407.16637 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://pod…

1
Course-Correction: Safety Alignment Using Synthetic Preferences 20:57

9m ago20:57

20:57

This paper evaluates and enhances large language models' ability to autonomously correct harmful content, introducing the C-EVAL benchmark and a synthetic dataset for effective preference learning. https://arxiv.org/abs//2407.16637 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://pod…

1
[QA] Solving The Travelling Salesman Problem Using A Single Qubit 8:03

31m ago8:03

8:03

This paper presents a resource-efficient quantum algorithm for solving the travelling salesman problem using a single qubit, achieving optimal solutions for small city samples with potential polynomial speed-up. https://arxiv.org/abs//2407.17207 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcast…

1
Solving The Travelling Salesman Problem Using A Single Qubit 21:52

31m ago21:52

21:52

This paper presents a resource-efficient quantum algorithm for solving the travelling salesman problem using a single qubit, achieving optimal solutions for small city samples with potential polynomial speed-up. https://arxiv.org/abs//2407.17207 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcast…

1
[QA] Exploring Scaling Trends in LLM Robustness 7:35

38m ago7:35

7:35

Larger language models show improved responses to adversarial training, but scaling alone does not enhance robustness against adversarial prompts without explicit defenses. https://arxiv.org/abs//2407.18213 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcas…

1
Exploring Scaling Trends in LLM Robustness 16:03

38m ago16:03

16:03

Larger language models show improved responses to adversarial training, but scaling alone does not enhance robustness against adversarial prompts without explicit defenses. https://arxiv.org/abs//2407.18213 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcas…

1
Benchmarking IR Models (w/ Nandan Thakur) 21:55

4h ago21:55

21:55

In this episode of Neural Search Talks, we're chatting with Nandan Thakur about the state of model evaluations in Information Retrieval. Nandan is the first author of the paper that introduced the BEIR benchmark, and since its publication in 2021, we've seen models try to hill-climb on the leaderboard, but also fail to outperform the BM25 baseline …

1
[QA] VILA2: VILA Augmented VILA 7:48

1d ago7:48

7:48

This paper presents VILA, a visual language model that enhances data quality and performance through self-augmentation and specialist-augmentation, achieving state-of-the-art results on various tasks. https://arxiv.org/abs//2407.17453 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://…

1
VILA2: VILA Augmented VILA 32:48

2h ago32:48

32:48

This paper presents VILA, a visual language model that enhances data quality and performance through self-augmentation and specialist-augmentation, achieving state-of-the-art results on various tasks. https://arxiv.org/abs//2407.17453 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://…

1
[QA] OpenDevin: An Open Platform for AI Software Developers as Generalist Agents 7:17

2h ago7:17

7:17

OpenDevin is a platform for developing AI agents that write code, interact with command lines, and browse the web, evaluated on 15 challenging tasks, fostering community contributions. https://arxiv.org/abs//2407.16741 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.c…

1
OpenDevin: An Open Platform for AI Software Developers as Generalist Agents 31:22

2h ago31:22

31:22

OpenDevin is a platform for developing AI agents that write code, interact with command lines, and browse the web, evaluated on 15 challenging tasks, fostering community contributions. https://arxiv.org/abs//2407.16741 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.c…

1
The Art of Dog Training With Francis Metcalf (Canine Circus School) 2:03:50

3h ago2:03:50

2:03:50

Grayson chats with on of his earliest dog training influences. Francis Metcalf is a man of diverse interests. A ringsport OG, pet trainer, trick training enthusiast, and artist, he has helped shaped the culture of American Dog training for more than three decades. Visit our sponsor: https://www.uglydoghunting.com Visit Grayson's Website: https://lo…

1
[QA] MovieDreamer: Hierarchical Generation for Coherent Long Visual Sequences 7:30

3h ago7:30

7:30

MovieDreamer is a novel framework combining autoregressive models and diffusion rendering for generating long-duration videos with complex narratives, high visual fidelity, and enhanced character consistency. https://arxiv.org/abs//2407.16655 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: …

1
MovieDreamer: Hierarchical Generation for Coherent Long Visual Sequences 24:34

3h ago24:34

24:34

MovieDreamer is a novel framework combining autoregressive models and diffusion rendering for generating long-duration videos with complex narratives, high visual fidelity, and enhanced character consistency. https://arxiv.org/abs//2407.16655 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: …

1
[QA] KAN or MLP: A Fairer Comparison 7:00

3h ago7:00

7:00

This paper compares KAN and MLP models across various tasks, finding MLP generally outperforms KAN, except in symbolic formula representation, where B-spline activation enhances performance. https://arxiv.org/abs//2407.16674 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.a…

1
KAN or MLP: A Fairer Comparison 21:31

3h ago21:31

21:31

This paper compares KAN and MLP models across various tasks, finding MLP generally outperforms KAN, except in symbolic formula representation, where B-spline activation enhances performance. https://arxiv.org/abs//2407.16674 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.a…

1
[QA] Conditioned Language Policy: A General Framework for Steerable Multi-Objective Finetuning 8:25

16m ago8:25

8:25

https://arxiv.org/abs//2407.15762 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016 Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers --- Support this podcast: https://podcasters.spotify.com/pod/show/arxiv-papers/supp…

1
Conditioned Language Policy: A General Framework for Steerable Multi-Objective Finetuning 31:53

16m ago31:53

31:53

https://arxiv.org/abs//2407.15762 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016 Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers --- Support this podcast: https://podcasters.spotify.com/pod/show/arxiv-papers/supp…

1
[QA] BOND: Aligning LLMs with Best-of-N Distillation 7:24

20m ago7:24

7:24

https://arxiv.org/abs//2407.14622 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016 Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers --- Support this podcast: https://podcasters.spotify.com/pod/show/arxiv-papers/supp…

1
BOND: Aligning LLMs with Best-of-N Distillation 25:45

20m ago25:45

25:45

https://arxiv.org/abs//2407.14622 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016 Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers --- Support this podcast: https://podcasters.spotify.com/pod/show/arxiv-papers/supp…

1
The Importance of Eye Contact During Dog Training 16:02

4h ago16:02

16:02

When acting selfishly while playing youth sports, the common phrase “there’s no I in team” was repeatedly thrown around. My coaches insisted I trust others with the ball to develop good teamwork. But giving up the ball didn’t come naturally. I often got “the eye” from my coaches, revealing their disappointment in my lack of team play. They’d use va…

1
Chukar guide dispels myths and dispenses tips; meet the Cesky Fousek 1:13:19

17m ago1:13:19

1:13:19

Is cheatgrass really the magic potion for chukar hunters? Pro chukar guide Grant Richie addresses some long-held notions about this bucket-list bird, from habitat and feed, to proven tactics and strategies. Grant runs hunts in the legendary Hell's Canyon region, some of the most rugged chukar terrain in the U.S. He offers tips and advice on finding…

1
[QA] Jumping Ahead: Improving Reconstruction Fidelity with JumpReLU Sparse Autoencoders 7:09

14m ago7:09

7:09

https://arxiv.org/abs//2407.14435 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016 Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers --- Support this podcast: https://podcasters.spotify.com/pod/show/arxiv-papers/supp…

1
Jumping Ahead: Improving Reconstruction Fidelity with JumpReLU Sparse Autoencoders 28:06

15m ago28:06

28:06

https://arxiv.org/abs//2407.14435 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016 Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers --- Support this podcast: https://podcasters.spotify.com/pod/show/arxiv-papers/supp…

1
[QA] LazyLLM: Dynamic Token Pruning for Efficient Long Context LLM Inference 7:25

19m ago7:25

7:25

LazyLLM accelerates transformer-based language model inference by dynamically selecting essential tokens for KV cache computation, improving generation speed without fine-tuning while maintaining accuracy across various tasks. https://arxiv.org/abs//2407.14057 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_paper…

1
LazyLLM: Dynamic Token Pruning for Efficient Long Context LLM Inference 23:05

20m ago23:05

23:05

LazyLLM accelerates transformer-based language model inference by dynamically selecting essential tokens for KV cache computation, improving generation speed without fine-tuning while maintaining accuracy across various tasks. https://arxiv.org/abs//2407.14057 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_paper…

1
Uncle Sam Explosions, Plane Engine Suckings, Carousel Squishings, and Elephant Crushings - ep. 216 40:55

1d ago40:55

40:55

On this, the 216th episode of the The Darwin Awards podcast, we discuss some of the awe-inspiring ways that people have recently eliminated themselves from the gene pool. These include a firework on a head, a classic plane engine story, a man rescuing a shoe, and someone trying to take a picture with an elephant. CLICK HERE TO SUBSCRIBE TO DARWIN A…

1
[QA] Scaling Laws with Vocabulary: Larger Models Deserve Larger Vocabularies 8:22

1h ago8:22

8:22

This study highlights the importance of vocabulary size in scaling large language models, proposing optimal sizes that enhance performance, particularly for models like Llama2-70B. https://arxiv.org/abs//2407.13623 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/u…

1
Scaling Laws with Vocabulary: Larger Models Deserve Larger Vocabularies 34:30

1h ago34:30

34:30

This study highlights the importance of vocabulary size in scaling large language models, proposing optimal sizes that enhance performance, particularly for models like Llama2-70B. https://arxiv.org/abs//2407.13623 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/u…

1
[QA] NeedleBench: Can LLMs Do Retrieval and Reasoning in 1 Million Context Window? 8:07

1h ago8:07

8:07

NeedleBench evaluates large language models' long-context capabilities, highlighting their struggles with logical reasoning in bilingual texts and suggesting improvements for practical applications. Resources are available at OpenCompass. https://arxiv.org/abs//2407.11963 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/…

1
NeedleBench: Can LLMs Do Retrieval and Reasoning in 1 Million Context Window? 25:18

1h ago25:18

25:18

NeedleBench evaluates large language models' long-context capabilities, highlighting their struggles with logical reasoning in bilingual texts and suggesting improvements for practical applications. Resources are available at OpenCompass. https://arxiv.org/abs//2407.11963 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/…

1
Training and Troubleshooting the Recall Command with Dogs 14:44

1d ago14:44

14:44

The recall is an extremely important ritual we have with our dogs. When called, the dog quickly and directly finds its position either in front or to our side. It happily and thoughtlessly finds its position, returning as excitedly as it went out and finishing with robotic consistency; something that has been groomed into it since the day it came h…

1
[QA] Q-Sparse: All Large Language Models can be Fully Sparsely-Activated 7:23

31m ago7:23

7:23

Q-Sparse is an efficient method for training sparsely-activated large language models, achieving comparable results to baseline models while significantly improving inference efficiency and reducing costs. https://arxiv.org/abs//2407.10969 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: htt…

1
Q-Sparse: All Large Language Models can be Fully Sparsely-Activated 21:57

32m ago21:57

21:57

Q-Sparse is an efficient method for training sparsely-activated large language models, achieving comparable results to baseline models while significantly improving inference efficiency and reducing costs. https://arxiv.org/abs//2407.10969 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: htt…

1
[QA] Foundational Autoraters: Taming Large Language Models for Better Automatic Evaluation 7:12

42m ago7:12

7:12

https://arxiv.org/abs//2407.10817 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016 Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers --- Support this podcast: https://podcasters.spotify.com/pod/show/arxiv-papers/supp…

1
Foundational Autoraters: Taming Large Language Models for Better Automatic Evaluation 38:59

43m ago38:59

38:59

https://arxiv.org/abs//2407.10817 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016 Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers --- Support this podcast: https://podcasters.spotify.com/pod/show/arxiv-papers/supp…

1
[QA] Scaling Retrieval-Based Language Models with a Trillion-Token Datastore 15:23

25m ago15:23

15:23

This paper explores how increasing datastore size enhances retrieval-based language models' performance, demonstrating that smaller models with large datastores outperform larger models in knowledge-intensive tasks. https://arxiv.org/abs//2407.12854 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Pod…

1
Scaling Retrieval-Based Language Models with a Trillion-Token Datastore 30:26

26m ago30:26

30:26

This paper explores how increasing datastore size enhances retrieval-based language models' performance, demonstrating that smaller models with large datastores outperform larger models in knowledge-intensive tasks. https://arxiv.org/abs//2407.12854 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Pod…

1
[QA] Beyond KV Caching: Shared Attention for Efficient LLMs 16:21

14m ago16:21

16:21

This paper presents a Shared Attention mechanism that improves the efficiency of large language models by sharing attention weights across layers, reducing computational resources while maintaining performance. https://arxiv.org/abs//2407.12866 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts…

1
Beyond KV Caching: Shared Attention for Efficient LLMs 23:14

33m ago23:14

23:14

This paper presents a Shared Attention mechanism that improves the efficiency of large language models by sharing attention weights across layers, reducing computational resources while maintaining performance. https://arxiv.org/abs//2407.12866 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts…

1
Episode 109 - our experience with immunotherapy and canine allergies 37:57

6h ago37:57

37:57

I talk here about my experience with immunotherapy for Ren's extreme allergies. Whilst there seem to be so many people around with itchy dogs, it seems that not many have explored immunotherapy and there isn't much information about it out there....By Jo Laurens

1
[QA] Private prediction for large-scale synthetic text generation 8:27

30m ago8:27

8:27

Approach for generating differentially private synthetic text using large language models through private prediction, enabling creation of thousands of high-quality data points for various applications. https://arxiv.org/abs//2407.12108 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https:…

1
Private prediction for large-scale synthetic text generation1 14:16

31m ago14:16

14:16

Approach for generating differentially private synthetic text using large language models through private prediction, enabling creation of thousands of high-quality data points for various applications. https://arxiv.org/abs//2407.12108 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https:…

1
[QA] Chip Placement with Diffusion 7:51

37m ago7:51

7:51

Novel diffusion model for macro placement in digital circuit design outperforms existing reinforcement learning methods by placing all components simultaneously. https://arxiv.org/abs//2407.12282 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-pap…

1
Chip Placement with Diffusion 12:16

37m ago12:16

12:16

Novel diffusion model for macro placement in digital circuit design outperforms existing reinforcement learning methods by placing all components simultaneously. https://arxiv.org/abs//2407.12282 YouTube: https://www.youtube.com/@ArxivPapers TikTok: https://www.tiktok.com/@arxiv_papers Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-pap…

Podcasts Worth a Listen

Retriever Training Podcasts

Podcasts Worth a Listen

Quick Reference Guide