arXiv:2611.00597v1 Announce Type: new Abstract: Does eliciting five weighted estimates within a single response change how a language model responds to information? Across anchoring, hindsight, base-rate, conjunction, and status-quo/default tasks, we compare direct numerical answers with a multi-es
arXiv:2611.00616v1 Announce Type: new Abstract: Inference-time scaling enhances a model's reasoning by extending its chain-of-thought (CoT). However, existing approaches typically rely on a single policy trained with outcome-reward reinforcement learning (RL), which often suffers from long-horizon
arXiv:2611.00638v1 Announce Type: new Abstract: Electrochemistry is central to solving many modern-day challenges, including energy storage systems for a sustainable future, and its advancement relies heavily on the data from past findings. Existing automated pipelines are limited by text-only arch
arXiv:2611.00665v1 Announce Type: new Abstract: Subliminal learning allows AI models to inherit hidden preferences from seemingly unrelated training data, such as number sequences. Safety audits may evaluate models at standard 16-bit precision, while deployments may use compressed, quantized weight
arXiv:2611.00672v1 Announce Type: new Abstract: When a language model decides refund or claim requests under written policies, the system can show it similar past cases. Each past case has two parts: its facts and the decision recorded for it. We separate the two. A model sees a policy, a request a
arXiv:2611.00726v1 Announce Type: new Abstract: KGW watermarks bias sampling to accumulate keyed detection evidence. We construct budgeted evasion through token-event evidence replacement using ordinary phrase rewrites. Our construction jointly selects compositions of LLM-proposed alternatives from
arXiv:2611.00755v1 Announce Type: new Abstract: Persistent Web information assistants must combine evolving document facts with user requirements established and revised across sessions. Yet systematic evaluation of how changes in both sources affect response requirements and evidence attribution r
arXiv:2611.00785v1 Announce Type: new Abstract: Vision-language models (VLMs) have shown promise for chest radiograph report generation, but adoption remains limited because of hallucinated findings that are not grounded in the image. Correcting these hallucinations and improving overall report qua
arXiv:2611.00796v1 Announce Type: new Abstract: Post training for large reasoning models remains insufficient on challenging problems, whose correct trajectories are hard to sample. Though prior works address this difficulty by integrating supervised fine tuning on expert trajectories with reinforc
arXiv:2611.00852v1 Announce Type: new Abstract: On-policy distillation (OPD), which uses reverse-KL supervision on student rollouts to transfer a context-conditioned model's behavior into student model parameters for closed-book inference, has shown strong results. However, we find that this succes
arXiv:2611.00871v1 Announce Type: new Abstract: Despite advances in retrieval-augmented generation (RAG), suppressing hallucinations during complex reasoning remains a persistent challenge. We identify that existing RAG pipelines suffer from a textual bottleneck: by passing only raw text to the gen
arXiv:2611.00877v1 Announce Type: new Abstract: Depth pruning removes whole Transformer blocks to make a language model smaller and faster, but the damage is not spread evenly across what the model can do, and a short fine-tuning run, called healing, decides what comes back. We measure this for sma
arXiv:2611.00883v1 Announce Type: new Abstract: Benchmark compression reduces the often prohibitive time and cost of model evaluation by evaluating models on a small benchmark subset that closely matches full-benchmark results. To quantify estimation error, we derive an accurate analytical approxim
arXiv:2611.00911v1 Announce Type: new Abstract: Language models are no longer used only to generate text. Software also asks them for decisions, in routing, guardrails, and automated gates. System One decision models, also called typed decision models, such as Jev and Laya, return one fast decision
arXiv:2611.00935v1 Announce Type: new Abstract: Advances in large language models (LLMs) have created opportunities for cross-lingual transfer of medical knowledge from high-resource to low-resource languages to support access to healthcare information. However, medical question-answering (QA) data
arXiv:2611.00939v1 Announce Type: new Abstract: Bock's algorithm provides a matrix-based alternative to Chu--Liu/Edmonds (CL/E) for exact dependency parsing. We present a reformulation that retains original node and arc identities through component labels, column adjustments, and star-and-bar trans
arXiv:2611.00960v1 Announce Type: new Abstract: Text-to-SQL assistants for streaming engines must handle dialect-specific types, windowing rules, and temporal semantics, including constructs poorly handled by generic prompts. We present Self-Refining Prompt Induction (SRPI), an offline method that
arXiv:2611.00983v1 Announce Type: new Abstract: Early-exit language models decide for each token how much computation to spend, usually by stopping once the prediction is confident. Yet a token is not a fixed unit: tokenizers cut words into different numbers of pieces in different languages. We ask
arXiv:2611.00986v1 Announce Type: new Abstract: Learning from extensive longitudinal clinical records--spanning repeated encounters, evolving conditions, and heterogeneous documentation--under a finite labeling budget requires solving two coupled problems: deciding what evidence to keep and decidin
arXiv:2611.01065v1 Announce Type: new Abstract: Gated knowledge distillation (KD) weights teacher signals per token or example, aiming to trust an imperfect teacher only where it seems reliable. We test two gates: EWAD (entropy-weighted agreement-aware distillation), which gates tokens by teacher c