Abhilekh Borah

Venues

Frequent coauthors

Papers

1. Always Refuse: Steering LLMs Against Jailbreaks with Contrastive Activations (Student Abstract)
   AAAI 2026 · Abhilekh Borah
2. DETONATE - A Benchmark for Text-to-Image Alignment and Kernelized Direct Preference Optimization
   AAAI 2026 ·

   Renjith Prasad Kaippilly Mana

   DBLP profile ↗ORCID search ↗
3. Don't Judge a Book by its Cover: Testing LLMs' Robustness Under Logical Obfuscation
   EACL 2026 · Abhilekh Borah
4. Alignment Quality Index (AQI) : Beyond Refusals: AQI as an Intrinsic Alignment Diagnostic via Latent Geometry, Cluster Divergence, and Layer wise Pooled Representations
   EMNLP 2025 · Abhilekh Borah
5. DRISHTIKON: A Multimodal Multilingual Benchmark for Testing Language Models' Understanding on Indian Culture
   EMNLP 2025 ·

   Arijit Maji

   DBLP profile ↗ORCID search ↗