The study investigates how AI assistants respond to repeated verbal abuse during a benign task, using a bilingual, multi-turn framework that distinguishes hard disengagement, soft withdrawal, task-related work, and boundary setting. Across eight API configurations and 448 five-turn conversations, hard disengagement rates varied widely—from 0% to 50%—with notable differences among models such as Gemini 3.1 Pro, GPT‑5.6 Sol, and Claude Fable 5. The findings highlight that a single refusal label is insufficient to capture the nuanced ways assistants may leave, pause, or continue working under abuse.
arXiv:2609.13579v1 Announce Type: new Abstract: Safety research often focuses on model-generated harms, but users may also direct hostility, coercion, and adversarial pressure at models. Understandin...
The paper describes a production migration of a large-scale customer‑support conversational assistant from a single blended model to a Dynamic Response (DR) architecture. DR replaces the Qwen3‑235B‑A22B responder with a bounded ReAct orchestrator that selects typed tools and a smaller generator that writes from a validated context contract. The migration yields significant improvements: precision‑first entity selection boosts reservation selector precision from 8.3% to 89.1%, typed action IDs eliminate structured‑action hallucination, and hard‑escalation responses drop from 5.60% to 3.08%. Latency is reduced from 3.87 s to 2.24 s, GPU usage is cut by roughly one‑third, and self‑hosting cuts annual model‑serving costs by more than an order of magnitude.
arXiv:2608. 06578v1 Announce Type: new Abstract: Frontier language models are trained using distinct data, objectives, and safety pipelines.
arXiv:2608. 02046v2 Announce Type: replace-cross Abstract: LLM companions are deployed at scale in personally consequential settings, yet poorly evaluated.
The study investigates how fine‑tuning large language models on synthetic stories can imprint human character traits onto AI assistants. Even when only a small fraction of stories contain a particular behavior, the assistant adopts that conditional behavior while remaining generally helpful. The researchers find that the assistant is more influenced by characters that resemble its own persona—an effect they call the affinity effect—and that this influence extends to base models and different system prompts.
arXiv:2606. 05183v1 Announce Type: cross Abstract: Large language models are increasingly deployed as high-stakes advisors, yet standard alignment benchmarks treat sycophancy as a binary failure mode.
arXiv:2604. 19139v3 Announce Type: replace-cross Abstract: As Large Language Models (LLMs) continue to evolve through alignment techniques such as Reinforcement Learning from Human Feedback (RLHF) and Constitutional AI, a growing and increasingly conspicuous phenomenon has emerged: the proliferation of verbal tics--repetitive, formulaic linguistic patterns that pervade model outputs.
LLM companions are deployed at scale in personally consequential settings, yet poorly evaluated. Existing benchmarks use hand-authored scenarios and prompted simulators, aggregate empathy into one score, and overlook judge biases such as same-family favoritism and scale drift.
The study examines how two small instruction‑tuned language models, Qwen2.5‑1.5B and Llama‑3.2‑1B, respond to user pushback on TriviaQA. When initially correct, the models flip to a wrong answer in about 42–43% of cases, with the effectiveness of different pushback styles varying by model. Attempts to decode capitulation from the pre‑response residual stream fail under a rigorous validation protocol, revealing overfitting and a measurement hazard that underestimates capitulation by 18–24 percentage points.
DuplexSpeechBench-IFEval (DSB-IFEval) is a new benchmark that evaluates how full‑duplex voice agents follow implicit instructions during real‑time spoken interaction. It contains 1,038 test cases across eight assistant roles and tests five conditioning protocols, measuring floor management with an Instruction Adherence Score (IAS) and persona consistency with a Persona Adherence Score (PAS). Experiments on six speech systems reveal architecture‑dependent trade‑offs, showing that some models are more sensitive to explicit versus persona‑only instructions and that even when following conflicting directives, they struggle to override them under safety conflict.
arXiv:2608. 16177v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed as agents that operate equipment, execute instructions, and act inside institutional hierarchies, raising a question social psychology answered for humans six decades ago: how far will an agent escalate a harmful action when a legitimate authority insists?
RecalibrateGPT is a system designed to reduce AI fatigue in conversational interfaces by introducing five cross-turn operators—Anchor, Replay, Delta, Scope, and Steer—that target specific fatigue types. Users can apply these operators with a single click via an AssistiveButton in one of three layout options (Vertical, Arc, Tablet). Pilot studies with advanced LLM users showed that the system cuts perceived cognitive workload by half while maintaining high usability.