Excess Description Length of Learning Generalizable Predictors

• URL: http://arxiv.org/abs/2601.04728v1
• Date: Thu, 08 Jan 2026 08:46:42 GMT
• Title: Excess Description Length of Learning Generalizable Predictors
• Authors: Elizabeth Donoway, Hailey Joren, Fabien Roger, Jan Leike,
• Abstract summary: We develop a formal information-theoretic framework for quantifying how much predictive structure fine-tuning extracts from a train dataset.<n>Our central quantity, Excess Description Length (EDL), is defined via prequential coding.<n>We establish that EDL is non-negative in expectation, converges to surplus description length in the infinite-data limit, and provides bounds on expected generalization gain.
• Score: 7.527535569795127
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Understanding whether fine-tuning elicits latent capabilities or teaches new ones is a fundamental question for language model evaluation and safety. We develop a formal information-theoretic framework for quantifying how much predictive structure fine-tuning extracts from the train dataset and writes into a model's parameters. Our central quantity, Excess Description Length (EDL), is defined via prequential coding and measures the gap between the bits required to encode training labels sequentially using an evolving model (trained online) and the residual encoding cost under the final trained model. We establish that EDL is non-negative in expectation, converges to surplus description length in the infinite-data limit, and provides bounds on expected generalization gain. Through a series of toy models, we clarify common confusions about information in learning: why random labels yield EDL near zero, how a single example can eliminate many bits of uncertainty about the underlying rule(s) that describe the data distribution, why structure learned on rare inputs contributes proportionally little to expected generalization, and how format learning creates early transients distinct from capability acquisition. This framework provides rigorous foundations for the empirical observation that capability elicitation and teaching exhibit qualitatively distinct scaling signatures.

Related papers

• Generalization Capability for Imitation Learning [1.30536490219656]
  Imitation learning holds the promise of equipping robots with versatile skills by learning from expert demonstrations.<n>However, policies trained on finite datasets often struggle to generalize beyond the training distribution.<n>We present a unified perspective on the generalization capability of imitation learning, grounded in both information theorey and data distribution property.
  arXiv  Detail & Related papers  (2025-04-25T17:59:59Z)
• I Predict Therefore I Am: Is Next Token Prediction Enough to Learn Human-Interpretable Concepts from Data? [76.15163242945813]
  Large language models (LLMs) have led many to conclude that they exhibit a form of intelligence.<n>We introduce a novel generative model that generates tokens on the basis of human-interpretable concepts represented as latent discrete variables.
  arXiv  Detail & Related papers  (2025-03-12T01:21:17Z)
• DISCO: DISCovering Overfittings as Causal Rules for Text Classification Models [6.369258625916601]
  Post-hoc interpretability methods fail to capture the models' decision-making process fully. Our paper introduces DISCO, a novel method for discovering global, rule-based explanations. DISCO supports interactive explanations, enabling human inspectors to distinguish spurious causes in the rule-based output.
  arXiv  Detail & Related papers  (2024-11-07T12:12:44Z)
• Accelerating Large Language Model Inference with Self-Supervised Early Exits [0.0]
  This paper presents a novel technique for accelerating inference in large, pre-trained language models (LLMs) We propose the integration of early exit ''heads'' atop existing transformer layers, which facilitate conditional terminations based on a confidence metric.
  arXiv  Detail & Related papers  (2024-07-30T07:58:28Z)
• Instruction Position Matters in Sequence Generation with Large Language Models [67.87516654892343]
  Large language models (LLMs) are capable of performing conditional sequence generation tasks, such as translation or summarization. We propose enhancing the instruction-following capability of LLMs by shifting the position of task instructions after the input sentences.
  arXiv  Detail & Related papers  (2023-08-23T12:36:57Z)
• Disentanglement via Latent Quantization [60.37109712033694]
  In this work, we construct an inductive bias towards encoding to and decoding from an organized latent space. We demonstrate the broad applicability of this approach by adding it to both basic data-re (vanilla autoencoder) and latent-reconstructing (InfoGAN) generative models.
  arXiv  Detail & Related papers  (2023-05-28T06:30:29Z)
• Large-scale Pre-trained Models are Surprisingly Strong in Incremental Novel Class Discovery [76.63807209414789]
  We challenge the status quo in class-iNCD and propose a learning paradigm where class discovery occurs continuously and truly unsupervisedly. We propose simple baselines, composed of a frozen PTM backbone and a learnable linear classifier, that are not only simple to implement but also resilient under longer learning scenarios.
  arXiv  Detail & Related papers  (2023-03-28T13:47:16Z)
• Harnessing the Power of Explanations for Incremental Training: A LIME-Based Approach [6.244905619201076]
  In this work, model explanations are fed back to the feed-forward training to help the model generalize better. The framework incorporates the custom weighted loss with Elastic Weight Consolidation (EWC) to maintain performance in sequential testing sets. The proposed custom training procedure results in a consistent enhancement of accuracy ranging from 0.5% to 1.5% throughout all phases of the incremental learning setup.
  arXiv  Detail & Related papers  (2022-11-02T18:16:17Z)
• An Explanation of In-context Learning as Implicit Bayesian Inference [117.19809377740188]
  We study the role of the pretraining distribution on the emergence of in-context learning. We prove that in-context learning occurs implicitly via Bayesian inference of the latent concept. We empirically find that scaling model size improves in-context accuracy even when the pretraining loss is the same.
  arXiv  Detail & Related papers  (2021-11-03T09:12:33Z)
• Masked Language Modeling and the Distributional Hypothesis: Order Word Matters Pre-training for Little [74.49773960145681]
  A possible explanation for the impressive performance of masked language model (MLM)-training is that such models have learned to represent the syntactic structures prevalent in NLP pipelines. In this paper, we propose a different explanation: pre-trains succeed on downstream tasks almost entirely due to their ability to model higher-order word co-occurrence statistics. Our results show that purely distributional information largely explains the success of pre-training, and underscore the importance of curating challenging evaluation datasets that require deeper linguistic knowledge.
  arXiv  Detail & Related papers  (2021-04-14T06:30:36Z)
• A Framework to Learn with Interpretation [2.3741312212138896]
  We present a novel framework to jointly learn a predictive model and its associated interpretation model. We seek for a small-size dictionary of high level attribute functions that take as inputs the outputs of selected hidden layers. A detailed pipeline to visualize the learnt features is also developed.
  arXiv  Detail & Related papers  (2020-10-19T09:26:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.