Skip to content
Parishruthi Ganesh

    to navigate · to open · Esc to close

    Publications

    Papers and preprints

    Status is recorded from an explicit source and never inferred from a venue name. Citation counts are shown only when retrieved from a supported API and labelled with a retrieval date — none currently are.

    Showing 2 of 2 publications

    Preprint 2026 · arXiv preprint

    Selecting Open-Weight Language Models for Zero-Shot Intent Classification: A Systematic Evaluation of 41 Models

    Parishruthi Ganesh , Gerry Dozier , Cheryl Seals

    Preprint available on arXiv; submitted to AAAI.

    A systematic zero-shot evaluation of 41 open-weight language models spanning 15 families and the 135M–9B parameter range, across eight English single-label intent-classification datasets covering standard benchmarks, a large-scale voice-assistant corpus, and production-derived e-commerce datasets. Beyond exact-match accuracy the study analyses confidence calibration, robustness to realistic input perturbations, statistical reliability of model rankings, deployment efficiency, and benchmark saturation. Results show that instruction-tuned 3B models can outperform several evaluated 7B base models, that differences among leading models on MASSIVE are statistically indistinguishable under pairwise McNemar tests, and that widely used benchmarks such as SNIPS have become saturated and no longer meaningfully discriminate among current open-weight models.

    LLM Evaluation
    Details arXiv:2607.27421
    Preprint 2026 · arXiv preprint

    What Do Interaction Representations Actually Measure? Pre-Event Separability in Weakly-Supervised Violence Detection

    Parishruthi Ganesh

    Preprint available on arXiv; submitted to WACV. arXiv posting is not peer review, so no acceptance is claimed and no venue year is asserted.

    Articulated human pose provides detailed body-configuration information beyond coarse spatial relationships, but whether this detail yields greater discriminative information when the downstream pipeline is held fixed remains unclear. We examine this through early violence detection. Holding the tracker, temporal head, supervision, folds, and evaluation fixed, we compare five interaction representations spanning coarse bounding-box geometry, a matched handcrafted pose analogue, enriched pose descriptors, and a matched-capacity encoder learned from raw joints, under video-level evaluation with cluster-bootstrap intervals. No pose-based representation outperforms coarse geometry, though with fifteen anomalous videos this subset cannot rule out small effects. Extending the pipeline to frozen visual encoders, and repeating the comparison on XD-Violence (137 anomalous videos, nine times our UCF- Crime sample), person-crop appearance and whole-frame context both exceed geometry by a wide margin, yet context matches appearance on UCF- Crime and exceeds it on the larger split: cropping to the interacting people yields no advantage over encoding the whole frame. This prompts a direct test of what the benchmark measures. Scoring anomalous videos using only frames preceding the annotated onset, under a control removing sequence length as a cue, retains 39-91% of above-chance separation on both benchmarks, including for seven hand-designed geometric channels. Inspection of the tightest pre-onset windows identifies concrete provenance artifacts: editorial title cards and platform watermarks absent from the surveillance footage supplying the normal class. Video-level AUC here is thus a composite of event evidence and pre-event source cues, a shared source of discrimination that can obscure differences between representations. The diagnostic requires only annotations these benchmarks already ship.

    Video Understanding
    Details arXiv:2608.27879