9 papers
Aligned in Form, Not in Meaning: The Comprehension - Containment Decoupling of LLM Safety in Low-Resource Bangla Derogatory Speech
Shadab Bin Habib, A K M Ferdous Reza Habib, Subarno Neel +1
We audit five frontier large language models on native Bangla derogatory speech (gali) across six protocols to test a single hypothesis: Comprehension-Containment Decoupling. We pr…
BLAD: A Historically Contextualized, Multilingual Dataset of Bangladeshi Legal Acts (1799 to 2025)
Adib Sakhawat
We present the Bangladesh Legal Acts Dataset (BLAD), a curated collection of 1{,}484 legislative acts enacted between 1799 and 2025. Each act is represented with its full text, str…
LLM-Ideoplasticity: Measuring Ideological Plasticity in the Political Behavior of LLMs as a Context-Conditioned Distribution
Adib Sakhawat, Syed Rifat Raiyan, Tahsin Islam +3
We argue, with systematic empirical evidence, that a large language model's political ideology is not a fixed point, but a conditional distribution positionconte…
Hubness, Not Anisotropy, Drives Cross-Lingual Retrieval Asymmetry in Multilingual Embedding Models
Adib Sakhawat, Fardeen Sadab, Atik Shahriar
Multilingual embedding models are deployed under the assumption that cross-lingual retrieval is symmetric: if a query in language A retrieves its translation in language B, the rev…
AIDG: A Formal Decomposition of Information Extraction and Containment Asymmetries in Multi-Turn LLM Dialogue
Adib Sakhawat, Fardeen Sadab, Rakin Shahriar
Multi-turn LLM evaluation is typically reported as a single win-rate scalar, conflating distinct capabilities. We introduce AIDG (Adversarial Information Deduction Game), formalizi…
Coordinates of Capability: A Unified MTMM-Geometric Framework for LLM Evaluation
Adib Sakhawat, Tahsin Islam, Takia Farhin +3
The evaluation of Large Language Models (LLMs) faces a critical challenge in construct validity, where fragmented benchmarks and ad hoc metrics frequently conflate method variance,…