collaborators

6 papers

cs.LG2026

CANARY: Zero-Label Detection of Fine-Tuning Contamination in Language Models

Swapnil Parekh

Adversaries can implant latent harmful behavior by poisoning as few as 1% of fine-tuning examples. The contamination is invisible to every output-level defense: harmful behavior li…

cs.LG2026

Drop the Act: Probe-Filtered RL for Faithful Chain-of-Thought Reasoning

Swapnil Parekh

Reasoning models post-hoc rationalize answers they have already committed to internally, producing chains of *reasoning theater*: deliberative-looking steps that contribute nothing…

cs.LG2026

Thinking Wrong in Silence: Backdoor Attacks on Continuous Latent Reasoning

Swapnil Parekh

A new generation of language models reasons entirely in continuous hidden states, producing no tokens and leaving no audit trail. We show that this silence creates a fundamentally…

cs.CL2026

CIRCUS: Circuit Consensus under Uncertainty via Stability Ensembles

Swapnil Parekh

Every mechanistic circuit carries an invisible asterisk: it reflects not just the model's computation, but the analyst's choice of pruning threshold. Change that choice and the cir…

cs.SD2026

ACES: Accent Subspaces for Coupling, Explanations, and Stress-Testing in Automatic Speech Recognition

Swapnil Parekh

ASR systems exhibit persistent performance disparities across accents, but whether these gaps reflect superficial biases or deep structural vulnerabilities remains unclear. We intr…

cs.CV2026

CaptionFool: Universal Image Captioning Model Attacks

Swapnil Parekh

Image captioning models are encoder-decoder architectures trained on large-scale image-text datasets, making them susceptible to adversarial attacks. We present CaptionFool, a nove…