Hallucinated — page 4

BrepCoder: A Unified Multimodal Large Language Model for Multi-task B-rep Reasoning (arxiv.org)

9h
Reinforcement Fine-Tuning of Flow-Matching Policies for Vision-Language-Action Models (arxiv.org)

9h fine-tuning
When are likely answers right? On Sequence Probability and Correctness in LLMs (arxiv.org)

9h
RolloutPipe: Overlapping Pipelined Rollout and Training in Disaggregated On-Policy LLM Reinforcement Learning (arxiv.org)

9h
Tailor Made Embeddings for Quantum Machine Learning (arxiv.org)

9h
Reinforcement Learning without Ground-Truth Solutions can Improve LLMs (arxiv.org)

9h
Hallucination in World Models is Predictable and Preventable (arxiv.org)

9h hallucination
- Hallucination in World Models Is Predictable and Preventable (www.nicklashansen.com via hn)
RSPC: A Benchmark for Modeling Stress and Psychiatric Conditions in Digitally Mediated Relationships using Psychiatrist Annotations (arxiv.org)

9h
Cross-Head Attention Uplift Network with Inverse Propensity Score under Unobserved Confounding (arxiv.org)

9h
Transformer-Based Classification of Bacterial Raman Spectra with LOOCV (arxiv.org)

9h
PersistentKV: Page-Aware Decode Scheduling for Long-Context LLM Serving on Commodity GPUs (arxiv.org)

9h
Empirical Software Engineering TerraProbe: A Layered-Oracle Framework for Detecting Deceptive Fixes in LLM-Assisted Terraform (arxiv.org)

9h
Can Large Language Models Reliably Code Qualitative Humanitarian Data? A Benchmark Study Against Human Expert Adjudication (arxiv.org)

9h
Optimizing CUDA like a Human: Micro-Profiling Tools as Expert Surrogates for LLM-Based GPU Kernel Optimization (arxiv.org)

9h
Rethinking Training & Inference for Forecasting: Linking Winner-Take-All back to GMMs (arxiv.org)

9h
At the Edge of Understanding: Sparse Autoencoders Trace The Limits of Transformer Generalization (arxiv.org)

9h
Dataset Usage Inference without Shadow Models or Held-out Data (arxiv.org)

9h
Necessary but Not Sufficient: Temperature Control and Reproducibility in LLM-as-Judge Safety Evaluations (arxiv.org)

9h
Spurious Rewards Paradox: Mechanistically Understanding How RLVR Activates Memorization Shortcuts in LLMs (arxiv.org)

9h
Eyes-on-Me: Scalable RAG Poisoning through Transferable Attention-Steering Attractors (arxiv.org)

9h rag
Not All Proofs Are Equal: Evaluating LLM Proof Quality Beyond Correctness (arxiv.org)

9h
Why Are Some Emotions Harder for LLMs? Uncovering the Causal Mechanisms of Emotion Inference via Sparse Autoencoders (arxiv.org)

9h
When Actions Go Off-Task: Detecting and Correcting Misaligned Actions in Computer-Use Agents (arxiv.org)

9h

Computer-use agents (CUAs) have made tremendous progress in the past year, yet they still frequently produce misaligned actions that deviate from the user's original intent. Such misaligned actions may arise from external attacks (e.g., in…
OI-Bench: An Option Injection Benchmark for Evaluating LLM Susceptibility to Directive Interference (arxiv.org)

9h

Benchmarking large language models (LLMs) is critical for understanding their capabilities, limitations, and robustness. In addition to interface artifacts, prior studies have shown that LLM decisions can be influenced by directive signals…
Overcoming State Inertia: Minimally Invasive Temporal Alignment for Evolving Contexts (arxiv.org)

9h dpo

Long-context dialogue systems suffer from state inertia, where models over-attend to history and fail to adapt to evolving intents. We demonstrate that standard alignment methods like DPO and even recent long-context optimization technique…
Vis-CoT: A Human-in-the-Loop Framework for Interactive Visualization and Intervention in LLM Chain-of-Thought Reasoning (arxiv.org)

9h

Large language models (LLMs) show strong reasoning via chain-of-thought (CoT) prompting, but the process is opaque, which makes verification, debugging, and control difficult in high-stakes settings. We present Vis-CoT, a human-in-the-loop…
Somatic in the East, Psychological in the West?: Investigating Clinically-Grounded Cross-Cultural Depression Symptom Expression in LLMs (arxiv.org)

9h

Prior clinical psychology research shows that Western individuals with depression tend to report psychological symptoms, while Eastern individuals report somatic ones. We test whether Large Language Models (LLMs), which are increasingly us…
The Geometry of Updates: Fisher Alignment at Vocabulary Scale (arxiv.org)

9h

Training-free source selection for LLM families with shared vocabularies arises in scientific string domains such as SMILES, protein, and genomic sequences, where candidate corpora share a tokenizer but differ in prediction targets. This c…
From Weights to Features: SAE-Guided Activation Regularization for LLM Continual Learning (arxiv.org)

9h

Weight-space regularization methods such as Elastic Weight Consolidation (EWC) are the standard approach to catastrophic forgetting in continual learning. However, those methods tend to underperform when applied to large language models.
Epiphany-Aware KV Cache Eviction Without the Attention Matrix (arxiv.org)

9h

As reasoning models emit chains of thought tens of thousands of tokens long, KV cache increasingly becomes a deployment bottleneck. Existing cache eviction methods rank tokens by attention weight, which is a noisy importance proxy in long…