Research map / Large language models / Language, safety and interpretability
Assistive Prompt Mediation: Evaluating Language Models Under Accessibility Constraints
Priyaranjan Pattnayak, Ishan Banerjee
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| increase impairment severity from alpha 0.2 to alpha >= 0.6 (ablation) | 10 frontier LLMs · APM intent preservation | IPR: mean IPR >= 4 at alpha=0.2, drops sharply at alpha >= 0.6 | Hurt |
| increase severity of N2 telegraphic noise from alpha 0.2 to 0.8 (ablation) | LLM mediators · APM | BRS (burden robustness): -5.81 -> -7.34 | Hurt |
| increase severity of N1 orthographic noise from alpha 0.2 to 0.8 (ablation) | LLM mediators · APM | BRS: 0.04 -> -0.39 | Hurt |
| mid-severity noise N2/N3 versus N1 orthographic noise | LLM mediators · APM | false robustness (high IPR with burden inflation) | Hurt |
| switch from English to Hindi prompts (ablation) | LLM mediators · APM | hallucination rate: 0.272 -> 0.396 | Hurt |
| switch from English to Chinese prompts (ablation) | LLM mediators · APM | hallucination rate: 0.272 -> 0.193 | Improved |
| use larger frontier models vs small models | LLM mediators · APM | hallucination rate: 0.452 (llama31-8b), 0.513 (ministral-3-3b) -> 0.183 (gpt-4o) | Improved |
| small model ministral-3-3b protocol compliance | ministral-3-3b · APM | PCR: PCR 0.826 but IPR 2.563 and HIR 0.513 | Mixed |
| judge with Gemini 3 Pro instead of GPT-5.2 (ablation) | APM evaluation · judge robustness | ranking correlation: rho approx 0.81-0.86 | No clear effect |
| measure usability (ABI, LCI) instead of intent preservation | APM severity curves · APM | median failure alpha: ABI and LCI fail at alpha 0.4; IPR at 0.6 | Mixed |
Closest papers by meaning
- Breaking Barriers: A Paradigm Shift in Technology Accessibility for Individuals with Physical Disabilities (AAAI 2025 · similarity 0.83)
- Position: Beyond Assistance – Reimagining LLMs as Ethical and Adaptive Co-Creators in Mental Health Care (ICML 2025 · similarity 0.83)
- SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal (ICLR 2025 · similarity 0.82)
- CARE: Decoding-Time Safety Alignment via Rollback and Introspection Intervention (NeurIPS 2025 · similarity 0.82)
- LifeEval: A Multimodal Benchmark for Assistive AI in Egocentric Daily Life Tasks (CVPR 2026 · similarity 0.81)
- Towards Scalable Web Accessibility Audit with MLLMs as Copilots (AAAI 2026 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
