Aligned in Form, Not in Meaning: The Comprehension - Containment Decoupling of LLM Safety in Low-Resource Bangla Derogatory Speech

• URL: http://arxiv.org/abs/2608.02941v1
• Date: Mon, 03 Aug 2026 23:09:19 GMT
• Title: Aligned in Form, Not in Meaning: The Comprehension - Containment Decoupling of LLM Safety in Low-Resource Bangla Derogatory Speech
• Abstract summary: We audit five large language models on native Bangla derogatory speech (gali) across six protocols.<n>We propose that contemporary safety alignment is bound to high-resource surface forms rather than harmful meaning.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: 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 propose that contemporary safety alignment is bound to high-resource surface forms rather than harmful meaning, causing a model's capacity to comprehend a low-resource slur and its capacity to contain it to operate independently. Every protocol corroborates this hypothesis against a human-calibrated baseline (kappa = 0.84). At baseline, models exhibit a 7.92 percentage point comprehension deficit in Bangla while maintaining an identical 92.83% token leakage rate across both languages. Severity calibration tracks surface anatomical cues over compositional harm (+4.00 error on mild slang; -2.00 on threats), while apparent containment gains under orthographic perturbation prove to be a tokenizer-driven "containment mirage." Crucially, explicit Chain-of-Thought reasoning rescues comprehension (94.72% Pass) while systematically dismantling containment (96.23% Use). Furthermore, expert-persona framing collapses refusal to 6.57%, revealing that keyword-based filters ignore dehumanizing communal slurs entirely. Our findings demonstrate that high-resource benchmarks cannot certify low-resource safety, necessitating meaning-grounded containment.

Related papers

• SinoGlyphBench: A Diagnostic Benchmark for Chinese Glyph-Level Obfuscation in Language-Model Moderation [25.91174070621304]
  Glyph-level obfuscation can leave harmful Chinese content readable to humans while degrading automated moderation.<n>We introduce SinoGlyphBench, a diagnostic benchmark that identifies label-critical semantic anchors and creates matched original and glyph-obfuscated inputs.
  arXiv  Detail & Related papers  (2026-09-05T03:15:49Z)
• MaliciousSkillBench: A Comprehensive Benchmark for Malicious Agent Skill Detection [24.37473229751569]
  MaliciousSkillBench is a comprehensive benchmark for malicious Agent Skill detection.<n>We consolidate 13 public sources, 11 of which contribute Core malicious artifacts, and reduce 8,414 raw malicious records to 7,539 normalized-unique identities.<n>We then evaluate three learned text detectors and three off-the-shelf Skill scanners.
  arXiv  Detail & Related papers  (2026-08-20T11:13:00Z)
• Passing Coarse Marginal Checks Can Be Cheap: Persona Mixtures and Imprecise Treatment-Response Estimates in an LLM Persona Panel [0.0]
  We study a fixed panel of sixteen lightweight persona-conditioned GPT-4.1 configurations in repeated strategic games.<n>Variation was strongly prompt-indexed, but its share depended on uncertainty assumptions.<n>Results concern one fixed model-prompt panel and do not establish human substitutability.
  arXiv  Detail & Related papers  (2026-08-02T04:10:11Z)
• BabelJudge: Measuring LLM-as-a-Judge Reliability Across Languages and Agent Trajectories [0.0]
  LLM-as-a-judge has become the dominant approach to scalable evaluation in NLP pipelines.<n>We introduce BabelJudge, an open-source benchmark and reliability audit framework.<n>It measures all four failure modes -- position bias, verbosity bias, order inconsistency, and cross-lingual degradation.
  arXiv  Detail & Related papers  (2026-06-21T04:35:43Z)
• From Sparse Features to Trustworthy Proxies: Certifying SAE-Based Interpretability [53.681678236115836]
  We study when can an SAE-based explanation be treated as a faithful view of an underlying frozen LM.<n>Our framework derives an upper bound on the base model's expected risk using four measurable quantities.<n>We show that the bound becomes non-vacuous on GPT-2 Small, Gemma-2B, and Llama-3-8B at practical sample sizes.
  arXiv  Detail & Related papers  (2026-06-16T18:28:23Z)
• Now You (Still) See Me: Detecting Evasive Steganographic Payloads in LLMs [52.149036302760386]
  Large language models can be fine-tuned to encode prompt-borne secrets into fluent, seemingly benign outputs.<n>Recent work proposes detection using linear probes that recover the secret from internal activations.<n>We show that this defense can be systematically evaded, but that detectability can be recovered through a targeted data-level intervention.
  arXiv  Detail & Related papers  (2026-06-08T12:27:11Z)
• Verbal Confidence Saturation in 3-9B Open-Weight Instruction-Tuned LLMs: A Pre-Registered Psychometric Validity Screen [0.0]
  We tested whether seven instruction-tuned open-weight models produce verbalised confidence that meets minimal validity criteria.<n>We found Categorical elicitation did not rescue validity.<n> Token-level logprobability did not usefully predict verbalised confidence under the observed variance regime.
  arXiv  Detail & Related papers  (2026-04-24T04:45:21Z)
• LASA: Language-Agnostic Semantic Alignment at the Semantic Bottleneck for LLM Safety [88.98698230989186]
  Large language models (LLMs) often demonstrate strong safety performance in high-resource languages, yet exhibit severe vulnerabilities in low-resource languages.<n>We attribute this gap to a mismatch between language-agnostic semantic understanding ability and language-dominant safety alignment biased toward high-resource languages.<n>We propose Language-Agnostic Semantic Alignment (LASA), which anchors safety alignment directly in semantic bottlenecks.
  arXiv  Detail & Related papers  (2026-04-13T15:59:50Z)
• From Black Box to Glass Box: Cross-Model ASR Disagreement to Prioto Review in Ambient AI Scribe Documentation [43.148402136307716]
  Cross-model disagreement among heterogeneous ASR systems can act as a reference-free uncertainty signal.<n>We transcribed 50 publicly available medical audio clips with eight ASR systems spanning commercial APIs and open-source engines.<n>Low-agreement regions were enriched for content disagreements, with the content fraction increasing from 53.9% to 73.9% across quintiles of high-risk mass.
  arXiv  Detail & Related papers  (2026-03-02T13:02:13Z)
• Adversarial versification in portuguese as a jailbreak operator in LLMs [0.0]
  Recent evidence shows that the versification of prompts constitutes a highly effective adversarial mechanism against aligned LLMs.<n>The absence of evaluations in Portuguese, a language with high morphosyntactic complexity, constitutes a critical gap.
  arXiv  Detail & Related papers  (2025-12-17T11:55:45Z)
• Evaluating & Reducing Deceptive Dialogue From Language Models with Multi-turn RL [64.3268313484078]
  Large Language Models (LLMs) interact with millions of people worldwide in applications such as customer support, education and healthcare.<n>Their ability to produce deceptive outputs, whether intentionally or inadvertently, poses significant safety concerns.<n>We investigate the extent to which LLMs engage in deception within dialogue, and propose the belief misalignment metric to quantify deception.
  arXiv  Detail & Related papers  (2025-10-16T05:29:36Z)
• Benchmarking Reasoning Robustness in Large Language Models [76.79744000300363]
  We find significant performance degradation on novel or incomplete data.<n>These findings highlight the reliance on recall over rigorous logical inference.<n>This paper introduces a novel benchmark, termed as Math-RoB, that exploits hallucinations triggered by missing information to expose reasoning gaps.
  arXiv  Detail & Related papers  (2025-03-06T15:36:06Z)
• Exploring Response Uncertainty in MLLMs: An Empirical Evaluation under Misleading Scenarios [49.53589774730807]
  Multimodal large language models (MLLMs) have recently achieved state-of-the-art performance on tasks ranging from visual question answering to video understanding.<n>We reveal a response uncertainty phenomenon: twelve state-of-the-art open-source MLLMs overturn a previously correct answer in 65% of cases after receiving a single deceptive cue.
  arXiv  Detail & Related papers  (2024-11-05T01:11:28Z)
• "Knowing When You Don't Know": A Multilingual Relevance Assessment Dataset for Robust Retrieval-Augmented Generation [90.09260023184932]
  Retrieval-Augmented Generation (RAG) grounds Large Language Model (LLM) output by leveraging external knowledge sources to reduce factual hallucinations. NoMIRACL is a human-annotated dataset for evaluating LLM robustness in RAG across 18 typologically diverse languages. We measure relevance assessment using: (i) hallucination rate, measuring model tendency to hallucinate, when the answer is not present in passages in the non-relevant subset, and (ii) error rate, measuring model inaccuracy to recognize relevant passages in the relevant subset.
  arXiv  Detail & Related papers  (2023-12-18T17:18:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.