CS-E407517 - Special Course in Machine Learning, Data Science and Artificial Intelligence: Seminar on NLP Research, Lectures, 4.9.2023-27.11.2023

Topic outline

• Presentation Schedule
  Session 2 (11.9.2023)

  Generating Images with Multimodal Language Models
  Presenter: Jasu Vehtari
  Opponent: Roope Kajoluoto

  Direct Preference Optimization: Your Language Model is Secretly a Reward Model
  Presenter: Ya Gao
  Opponent: Katja Voskoboinik
  

  Session 3 (18.9.2023)

  Training Compute-Optimal Large Language Models
  Presenter: Severi Rissanen
  Opponent: Jatan Shrestha
  

  Self-instruct: Aligning Language Models with Self-Generated Instructions
  Presenter: Matteo Merler
  Opponent: Miquel Marin
  

  Session 4 (25.9.2023)

  Parameter-Efficient Transfer Learning for NLP
  Presenter: Xinyu Zhang
  Opponent: Rafael Leal
  

  LoRA: Low-Rank Adaptation of Large Language Models
  Presenter: Paul Chang
  Opponent: Egor Eremin
  

  Session 5 (2.10.2023)

  LIMA: Less Is More for Alignment
  

  Presenter: Katja Voskoboinik
  

  Opponent: Paul Chang
  
  

  RoFormer: Enhanced Transformer with Rotary Position Embedding
  Presenter: Egor Eremin
  Opponent: Riccardo Mereu
  

  Session 6 (9.10.2023)

  Data Distributional Properties Drive Emergent In-Context Learning in Transformers

  Presenter: Hieu Nguyen
  Opponent: Ya Gao

  In-Context Learning in Large Language Models Learns Label Relationships but Is Not Conventional Learning
  Presenter: Miquel Marin
  Opponent: Blerta Lindqvist

  No session on 16.10. (Exam Week)

  
  Session 7 (23.10.2023)

  Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
  Presenter: Jatan Shrestha
  Opponent: Ali Salloum

  
  

  Tree of Thoughts: Deliberate Problem Solving with Large Language Models
  

  Presenter: Hoanh Le
  Opponent: Charles Koutcheme
  

  
  

  Session 8 (30.10.2023)

  Learning Transferable Visual Models From Natural Language Supervision
  

  Presenter: Blerta Lindqvist
  Opponent: Xinyu Zhang
  

  
  

  BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models

  Presenter: Roope Kajoluoto
  Opponent: Xu Ruixin
  

  
  

  Session 9 (6.11.2023)

  Evaluating Large Language Models Trained on Code
  Presenter: Charles Koutcheme
  Opponent: Jasu Vehtari
  

  
  

  Generative Agents: Interactive Simulacra of Human Behavior
  

  Presenter: Ali Salloum
  Opponent: Matteo Merler

  
  

  Session 10 (13.11.2023)

  Causal Parrots: Large Language Models May Talk Causality But Are Not Causal
  Presenter: Nazaal Ibrahim
  Opponent: Severi Rissanen
  

  
  

  Retentive Network: A Successor to Transformer for Large Language Models

  Presenter: Xu Ruixin
  Opponent: Hieu Nguyen

  
  

  Session 11 (20.11.2023)

  FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness
  

  FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning

  Presenter: Riccardo Mereu

  Opponent: Nazaal Ibrahim

  
  

  Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback

  Presenter: Rafael Leal
  Opponent: Hoanh Le