mlsanityData to 7 Oct 2026

Layer by Layer: Uncovering Where Multi-Task Learning Happens in Instruction-Tuned Large Language Models

arXiv 2410.20008 ↗First posted 25 Oct 2024cs.CL cs.LGcited by 15

Not on the map. Why?

Abstract (arXiv)

Fine-tuning pre-trained large language models (LLMs) on a diverse array of tasks has become a common approach for building models that can solve various natural language processing (NLP) tasks. However, where and to what extent these models retain task-specific knowledge remains largely unexplored. This study investigates the task-specific information encoded in pre-trained LLMs and the effects of instruction tuning on their representations across a diverse set of over 60 NLP tasks. We use a set of matrix analysis tools to examine the differences between the way pre-trained and instruction-tuned LLMs store task-specific information. Our findings reveal that while some tasks are already encoded within the pre-trained LLMs, others greatly benefit from instruction tuning. Additionally, we pinpointed the layers in which the model transitions from high-level general representations to more task-oriented representations. This finding extends our understanding of the governing mechanisms of LLMs and facilitates future research in the fields of parameter-efficient transfer learning and multi-task learning.

Key facts. Cited by 15 papers in the map (data through 7 Oct 2026). Builds on 23 of its 37 references that are in the map. First posted on arXiv 25 Oct 2024.

Builds on

23 of 37 references.

References · 23

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  4. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  5. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  6. 2020Measuring Massive Multitask Language Understandingcited by 7,768
  7. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,304
  8. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,747
  9. 2022Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Themcited by 1,767
  10. 2019Similarity of Neural Network Representations Revisitedcited by 1,695
  11. 2020A Primer in BERTology: What we know about how BERT workscited by 807
  12. 2023The Flan Collection: Designing Data and Methods for Effective Instruction Tuningcited by 734
11 more
  1. 2021Cross-Task Generalization via Natural Language Crowdsourcing Instructionscited by 696
  2. 2023Improving Text Embeddings with Large Language Modelscited by 295
  3. 2021Muppet: Massive Multi-task Representations with Pre-Finetuningcited by 192
  4. 2017What do Neural Machine Translation Models Learn about Morphology?cited by 165
  5. 2020What Happens To BERT Embeddings During Fine-tuning?cited by 113
  6. 2018Under the Hood: Using Diagnostic Classifiers to Investigate and Improve how Language Models Track Agreement Informationcited by 97
  7. 2021First Align, then Predict: Understanding the Cross-Lingual Ability of Multilingual BERTcited by 64
  8. 2018Evaluating Layers of Representation in Neural Machine Translation on Part-of-Speech and Semantic Tagging Taskscited by 56
  9. 2024Spectral Editing of Activations for Large Language Model Alignmentcited by 54
  10. 2023A Joint Matrix Factorization Analysis of Multilingual Representationscited by 5
  11. 2022Understanding Domain Learning in Language Models Through Subpopulation Analysiscited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 6; 2026: 9.

15 papers.

Sort:

By area:

  1. 2025Layer by Layer: Uncovering Hidden Representations in Language Modelscited by 293
  2. 2025Encode, Think, Decode: Scaling test-time reasoning with recursive latent thoughtscited by 23
  3. 2025Alignment Quality Index (AQI) : Beyond Refusals: AQI as an Intrinsic Alignment Diagnostic via Latent Geometry, Cluster Divergence, and Layer wise Pooled Representationscited by 7
  4. 2026Old Habits Die Hard: How Conversational History Geometrically Traps LLMscited by 5
  5. 2025FLoRIST: Singular Value Thresholding for Efficient and Accurate Federated Fine-Tuning of Large Language Modelscited by 4
  6. 2026Differentiable Faithfulness Alignment for Cross-Model Circuit Transfercited by 3
  7. 2025Parameter Importance-Driven Continual Learning for Foundation Modelscited by 2
  8. 2026Delve into the Applicability of Advanced Optimizers for Multi-Task Learningcited by 2
  9. 2026Pseudo-Unification: Entropy Probing Reveals Divergent Information Patterns in Unified Multimodal Modelscited by 1
  10. 2026Self Knowledge Re-expression: A Fully Local Method for Adapting LLMs to Tasks Using Intrinsic Knowledgecited by 1
  11. 2026On the Persistent Effects of Lexicality in Large Language Modelscited by 1
  12. 2025Multi Attribute Bias Mitigation via Representation Learningcited by 0
3 more
  1. 2026The Horizon Threshold in Cooperative Multi-Agent Reward-Free Explorationcited by 0
  2. 2026The Convergence Gap: Instruction-Tuned Language Models Stabilize Later in the Forward Passcited by 0
  3. 2026Training Prompt Matters: State-Adaptive Optimization for Robust Fine-Tuningcited by 0