Uploaded February 2024 | Updated September 2026, 53 minutes ago
Hey everyone! Thank you so much for watching the 87th episode of the Weaviate Podcast! I am SUPER excited to welcome Karel D'Oosterlinck! Karel is the creator of IReRa (Infer-Retrieve-Rank)! IReRa is one of the most impressive systems that have been built for Extreme Multi-Label Classification, leveraging the emerging paradigm of DSPy compilation! This podcast dives into all things IReRa, XMC, DSPy compilation, and applications in Biomedical NLP and Recommendation! I hope you find this useful!
XMC.dspy: github.com/KarelDO/xmc.dspy
Infer-Retrieve-Rank: arxiv.org/pdf/2401.12178.pdf
Karel D'Oosterlinck on X: twitter.com/KarelDoostrlnck
DSPy: github.com/stanfordnlp/dspy/tree/main
DSPy arxiv: arxiv.org/pdf/2310.03714.pdf
Chapters
0:00 Welcome Karel!
0:33 Extreme Multi-Label Classification
2:10 IReRa - How can LLMs do this?
9:10 The DSPy Compiler
12:25 IReRa Optimization
21:30 Input-Dependent Examples
24:48 Knowledge in Prompts
26:06 Model Compression - Ollama vs. GPT-4
28:38 The DSPy Signature Optimizer
31:07 OUTPUT JSON, or else…
36:28 Teacher-Student LLM Optimization
48:34 Designing LLM Programs
51:58 Ensembling LLMs
58:50 IReRa for Recommendation
1:05:00 Biomedical NLP
Hey everyone! Thank you so much for watching the 87th episode of the Weaviate Podcast! I am SUPER excited to welcome Karel D'Oosterlinck! Karel is the creator of IReRa (Infer-Retrieve-Rank)! IReRa is one of the most impressive systems that have been built for Extreme Multi-Label Classification, leveraging the emerging paradigm of DSPy compilation! This podcast dives into all things IReRa, XMC, DSPy compilation, and applications in Biomedical NLP and Recommendation! I hope you find this useful!
XMC.dspy: github.com/KarelDO/xmc.dspy
Infer-Retrieve-Rank: arxiv.org/pdf/2401.12178.pdf
Karel D'Oosterlinck on X: twitter.com/KarelDoostrlnck
DSPy: github.com/stanfordnlp/dspy/tree/main
DSPy arxiv: arxiv.org/pdf/2310.03714.pdf
Chapters
0:00 Welcome Karel!
0:33 Extreme Multi-Label Classification
2:10 IReRa - How can LLMs do this?
9:10 The DSPy Compiler
12:25 IReRa Optimization
21:30 Input-Dependent Examples
24:48 Knowledge in Prompts
26:06 Model Compression - Ollama vs. GPT-4
28:38 The DSPy Signature Optimizer
31:07 OUTPUT JSON, or else…
36:28 Teacher-Student LLM Optimization
48:34 Designing LLM Programs
51:58 Ensembling LLMs
58:50 IReRa for Recommendation
1:05:00 Biomedical NLP
