arXiv AI
When Should Enterprise Policy Decisions Be Learned, and When Should They Be Reasoned?
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv AI
Jul 29
Measuring and Improving Behavioral Consistency in Large Language Models through Fact-Heuristic-Emotion State Enforcement
arXiv:2607. 24765v1 Announce Type: cross Abstract: Large language models (LLMs) can give different answers to the same decision problem across runs, and reverse a decision when their own prior answer returns as context.
arXiv Computation and Language
2d ago
Precedent Labels and Facts Shift LLM Approval Decisions in Opposite Directions
The study investigates how language models (LLMs) handle refund or claim requests under written policies when presented with past cases that include both facts and recorded decisions. By separating these two components, the authors found that changing the recorded decisions of near‑copy past cases can increase approval rates by 30–55 percentage points, while altering the facts to support approval actually decreases approval rates, suggesting that the request appears weaker when compared to such facts. These effects were consistent across multiple models and both study‑specific and publicly paraphrased policies.
arXiv AI
Jul 21
Committed Before Reasoning: Behavioral Reproduction and Preliminary Activation-Level Evidence of Answer Pre-Commitment in an Open-Weight LLM
arXiv:2607. 16451v1 Announce Type: cross Abstract: Chat models sometimes commit to an answer and then produce reasoning that justifies it rather than deriving it -- even when the answer contradicts a task premise.
arXiv AI
Sep 12
Off-Target Effects of Response-Style Alignment in a Korean 27B Language Model
arXiv:2609. 11291v1 Announce Type: new Abstract: We post-train Qwen3.
arXiv AI
Oct 27
Trainable Boundary-Keeping Where No Formal Answer Exists: A Controlled Knock-In Study of Reward Training in Evidence-Sufficiency Judgment
The study evaluates whether reward‑trained language models can learn to keep evidence‑sufficiency boundaries without a formal correctness signal. Using a controlled knock‑in design, a boundary‑keeping recipe (SFT followed by GRPO) dramatically increased spontaneous correct refusals from 25 % to 100 %, while a scrambled‑pairing placebo collapsed performance. The results show that final boundary‑keeping is trainable without an inference‑time verifier, though active numeric checking remains negligible.
arXiv Machine Learning
Oct 15
ActiveEvidence: Decision-Focused Evidence Graphs with Verifiable Acquisition Gaps
arXiv:2610.20532v1 Announce Type: cross Abstract: A multimodal agent must decide whether visual evidence is worth acquiring, which candidate or resolution to query, and when further interaction can n...