Decoupling Perception from Description: Computation-Grounded Representation Alignment between Multivariate Time Series and Language
Training multimodal models to align time series with language runs into a self-supervision trap. The usual recipe asks an LLM to read a series and write a description, so label quality is capped by the perceptual skill the model is supposed to learn. The data can never teach more than the labeler already knows. A second gap makes this worse: most datasets use a single variable, but the patterns that matter (cross-channel correlation, lead-lag structure, co-occurring anomalies) appear only with s
Record details
Published: 5 August 2026
Source: arXiv cs.LG
Category: Research
Topics: Safety & alignment
Retrieved: 7 August 2026
Related evidence
These records share source-supplied organisations, an exact publisher byline, automatic topics or regions. The reason is shown on every link; related does not mean supporting, agreeing with or verifying this record.
RepoProbe: Benchmarking Architecture-Aware Repository Comprehension with Checklists
arXiv cs.AI · 5 August 2026
Item Response Theory for AI Safety
arXiv · 5 August 2026
Agent Against Agent: An Agentic System for Automatic Prompt Injection Red Teaming
arXiv red teaming query · 5 August 2026
Rectifying Geometric Misalignment: Online Source-Free Adaptation for Class-Imbalanced EEG
arXiv cs.LG · 5 August 2026
Scaling Inherently Interpretable Language Models
HuggingFace Daily Papers · 5 August 2026
From Continuous Predictors to Clinical Thresholds: Early Evidence on Performance Trade-offs of Guideline-Based Categorisation for Ischaemic Stroke Outcome Prediction
arXiv cs.LG · 5 August 2026
How to cite this record
ethics.ai (5 August 2026), “Decoupling Perception from Description: Computation-Grounded Representation Alignment between Multivariate Time Series and Language,” evidence record 17397, https://ethics.ai/record/17397 (originally published by arXiv cs.LG).
Use and limitations
This page is a stable index and citation surface for a source record. ethics.ai did not author the underlying report and has not independently verified every claim. Automatic topics may be imperfect. For consequential use, quote and cite the original publisher.