From Generalist to Specialist Representation

• URL: http://arxiv.org/abs/2605.12733v1
• Date: Tue, 12 May 2026 20:34:53 GMT
• Title: From Generalist to Specialist Representation
• Abstract summary: Given a generalist model, learning a task-relevant specialist representation is fundamental for downstream applications.<n>We prove that the structure between time steps and tasks is identifiable in a fully unsupervised manner.<n>We then prove that, within each time step, the task-relevant latent representation can be disentangled from the irrelevant part.
• Score: 35.95654356089404
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Given a generalist model, learning a task-relevant specialist representation is fundamental for downstream applications. Identifiability, the asymptotic guarantee of recovering the ground-truth representation, is critical because it sets the ultimate limit of any model, even with infinite data and computation. We study this problem in a completely nonparametric setting, without relying on interventions, parametric forms, or structural constraints. We first prove that the structure between time steps and tasks is identifiable in a fully unsupervised manner, even when sequences lack strict temporal dependence and may exhibit disconnections, and task assignments can follow arbitrarily complex and interleaving structures. We then prove that, within each time step, the task-relevant latent representation can be disentangled from the irrelevant part under a simple sparsity regularization, without any additional information or parametric constraints. Together, these results establish a hierarchical foundation: task structure is identifiable across time steps, and task-relevant latent representations are identifiable within each step. To our knowledge, each result provides a first general nonparametric identifiability guarantee, and together they mark a step toward provably moving from generalist to specialist models.

Related papers

• Partial Identification under Causal Orders by Linear Programming [46.16293902963101]
  We show that any counterfactual inquiry induces a partial, topological ordering over relevant variables.<n>This allows bounding arbitrary counterfactual and nested counterfactual queries.
  arXiv  Detail & Related papers  (2026-08-25T11:42:38Z)
• Self-Certification of Representation Adequacy: Sequential Certification at Minimum Task Loss [4.6304326752276985]
  This paper develops a four-layer theory of self-certification of representation adequacy.<n>The sequential layer poses certification as an optimal-stopping problem in the currency of task loss.<n>A final boundary layer identifies the open theorem needed to cover policy switching or representation repair.
  arXiv  Detail & Related papers  (2026-08-03T14:09:34Z)
• Bridging Data Trials and Task Barriers: A Unified Framework for Sketch Biometric Identification [99.67657930163358]
  Sketch biometric identification faces challenges, including scarce real data, high annotation costs, privacy risks, and insufficient generalization ability of cross-task models.<n>This paper proposes a unified framework that integrates efficient synthetic sketch generation and task-sequential continual learning.<n>To support the study of sketch biometric identification, we built a new large-scale benchmark, Sketch-BioID, with several practical evaluation protocols.
  arXiv  Detail & Related papers  (2026-05-17T10:20:57Z)
• PRISM: : Planning and Reasoning with Intent in Simulated Embodied Environments [59.07829883257003]
  Built on five multi-room apartments, PRISM structures 300 human-verified tasks into three capability tiers.<n>PRISM exposes an agent-agnostic executable action API that allows arbitrary agents to be evaluated end-to-end.
  arXiv  Detail & Related papers  (2026-05-12T04:59:47Z)
• How LLMs Follow Instructions: Skillful Coordination, Not a Universal Mechanism [2.5997274006052544]
  We investigate whether instruction-following relies on a universal mechanism or compositional skill deployment.<n>General probes trained across all tasks consistently underperform task-specific specialists, indicating limited representational sharing.<n> temporal analysis shows constraint satisfaction operates as dynamic monitoring during generation rather than pre-generation planning.
  arXiv  Detail & Related papers  (2026-04-07T16:12:52Z)
• Detecting and Identifying Selection Structure in Sequential Data [53.24493902162797]
  We argue that the selective inclusion of data points based on latent objectives is common in practical situations, such as music sequences. We show that selection structure is identifiable without any parametric assumptions or interventional experiments. We also propose a provably correct algorithm to detect and identify selection structures as well as other types of dependencies.
  arXiv  Detail & Related papers  (2024-06-29T20:56:34Z)
• Leveraging Task Structures for Improved Identifiability in Neural Network Representations [31.863998589693065]
  We extend the theory of identifiability in supervised learning by considering the consequences of having access to a distribution of tasks. We show that linear identifiability is achievable in the general multi-task regression setting.
  arXiv  Detail & Related papers  (2023-06-26T17:16:50Z)
• Leveraging sparse and shared feature activations for disentangled representation learning [112.22699167017471]
  We propose to leverage knowledge extracted from a diversified set of supervised tasks to learn a common disentangled representation. We validate our approach on six real world distribution shift benchmarks, and different data modalities.
  arXiv  Detail & Related papers  (2023-04-17T01:33:24Z)
• Generalized Inverse Planning: Learning Lifted non-Markovian Utility for Generalizable Task Representation [83.55414555337154]
  In this work, we study learning such utility from human demonstrations. We propose a new quest, Generalized Inverse Planning, for utility learning in this domain. We outline a computational framework, Maximum Entropy Inverse Planning (MEIP), that learns non-Markovian utility and associated concepts in a generative manner.
  arXiv  Detail & Related papers  (2020-11-12T21:06:26Z)
• Structure by Architecture: Structured Representations without Regularization [31.75200752252397]
  We study the problem of self-supervised structured representation learning using autoencoders for downstream tasks such as generative modeling. We design a novel autoencoder architecture capable of learning a structured representation without the need for aggressive regularization. We demonstrate how these models learn a representation that improves results in a variety of downstream tasks including generation, disentanglement, and extrapolation.
  arXiv  Detail & Related papers  (2020-06-14T04:37:08Z)

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.