# More Features Are Not More Evidence: Limits of Training-Free Human Activity Recognition with Jev

arXiv: https://arxiv.org/abs/2609.36154 · HTML: https://mlsanity.com/paper/2609.36154/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 25 references that are in the map.
- First posted on arXiv 28 Sep 2026.

## Builds on (most cited first, 11 of 11)

- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [Selective Classification for Deep Neural Networks](https://mlsanity.com/paper/1705.08500/) (2017)
- [Deep Learning for Sensor-based Activity Recognition: A Survey](https://mlsanity.com/paper/1707.03502/) (2017)
- [ZARA: Training-Free Motion Time-Series Reasoning via Evidence-Grounded LLM Agents](https://mlsanity.com/paper/2508.04038/) (2025)
- [Calibrated Decisions at Scale: Converting Police Crash Narratives into Probabilistic Crash Variables with a System One Model (Jev)](https://mlsanity.com/paper/2609.24052/) (2026)
- [Large Language Models are Zero-Shot Recognizers for Activities of Daily Living](https://mlsanity.com/paper/2407.01238/) (2024)
- [Can Jev Judge Radiology Reports? Evaluating a System One Model for Clinical Factuality](https://mlsanity.com/paper/2609.27607/) (2026)
- [Foundation Models Defining A New Era In Sensor-based Human Activity Recognition: A Survey And Outlook](https://mlsanity.com/paper/2604.02711/) (2026)
- [JEVQA - Video Quality from Metadata, Bitstream, and Pixel Features with a General-Purpose Decision Model](https://mlsanity.com/paper/2609.24395/) (2026)
- [Large Language Models Memorize Sensor Datasets! Implications on Human Activity Recognition Research](https://mlsanity.com/paper/2406.05900/) (2024)
- [RAG-HAR+: Towards Cost-Efficient LLM-Based Human Activity Recognition for Edge Deployment](https://mlsanity.com/paper/2607.26631/) (2026)

## Abstract (arXiv)

General-purpose models promise sensor-based decisions without training a task-specific classifier, which could reduce the dependence of Human Activity Recognition (HAR) on labeled data. Yet it remains unclear whether such models can directly interpret deterministic descriptions of physical sensor signals well enough to replace or complement trained HAR models. We study this question using Jev, a fixed general-purpose probabilistic decision model, on 1,800 class-balanced accelerometer windows from WISDM, UCI341, and PAMAP2. Jev receives no labeled examples, retrieval context, or HAR-specific parameter updates. We evaluate three deterministic sensor representations and compare 5,400 Jev decisions with a generative baseline and three supervised HAR models. Jev remains far below supervised recognition, with its strongest representation reaching macro-F1 of 0.038, 0.118, and 0.089 across the three datasets, compared with 0.686 to 0.907 for the supervised models. More numerical features do not improve Jev. Instead, they reduce recognition on all three datasets, while augmenting the same numerical evidence with a deterministic semantic rendering partially recovers performance, although the experiment does not isolate semantics from the accompanying serialization and redundancy changes. Jev is fast and inexpensive to query, but its probabilities are not reliably calibrated for recognition. A post-hoc fusion analysis finds a small improvement on WISDM that does not replicate on UCI341 or PAMAP2. These results show that training-free sensor decisions depend not only on the information available in the signal, but also on whether the model can use the representation through which that information is exposed. The sensor-to-model interface should therefore be treated as part of the model evaluation rather than as a neutral preprocessing step.