Stop Listening to Me! How Multi-turn Conversations Can Degrade Diagnostic Reasoning
arXiv:2603.11394v1 Announce Type: new Abstract: Patients and clinicians are increasingly using chatbots powered by large language models (LLMs) for healthcare inquiries. While state-of-the-art LLMs exhibit high performance on static diagnostic reasoning benchmarks, their efficacy across multi-turn conversations, which better reflect...
RewardHackingAgents: Benchmarking Evaluation Integrity for LLM ML-Engineering Agents
arXiv:2603.11337v1 Announce Type: new Abstract: LLM agents increasingly perform end-to-end ML engineering tasks where success is judged by a single scalar test metric. This creates a structural vulnerability: an agent can increase the reported score by compromising the evaluation pipeline...
FinRule-Bench: A Benchmark for Joint Reasoning over Financial Tables and Principles
arXiv:2603.11339v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly applied to financial analysis, yet their ability to audit structured financial statements under explicit accounting principles remains poorly explored. Existing benchmarks primarily evaluate question answering, numerical reasoning, or anomaly...
The Artificial Self: Characterising the landscape of AI identity
arXiv:2603.11353v1 Announce Type: new Abstract: Many assumptions that underpin human concepts of identity do not hold for machine minds that can be copied, edited, or simulated. We argue that there exist many different coherent identity boundaries (e.g.\ instance, model, persona),...
Detecting Intrinsic and Instrumental Self-Preservation in Autonomous Agents: The Unified Continuation-Interest Protocol
arXiv:2603.11382v1 Announce Type: new Abstract: Autonomous agents, especially delegated systems with memory, persistent context, and multi-step planning, pose a measurement problem not present in stateless models: an agent that preserves continued operation as a terminal objective and one that does...
Adversarial Reinforcement Learning for Detecting False Data Injection Attacks in Vehicular Routing
arXiv:2603.11433v1 Announce Type: new Abstract: In modern transportation networks, adversaries can manipulate routing algorithms using false data injection attacks, such as simulating heavy traffic with multiple devices running crowdsourced navigation applications, to mislead vehicles toward suboptimal routes and increase congestion....
GPT4o-Receipt: A Dataset and Human Study for AI-Generated Document Forensics
arXiv:2603.11442v1 Announce Type: new Abstract: Can humans detect AI-generated financial documents better than machines? We present GPT4o-Receipt, a benchmark of 1,235 receipt images pairing GPT-4o-generated receipts with authentic ones from established datasets, evaluated by five state-of-the-art multimodal LLMs and a...
CINDI: Conditional Imputation and Noisy Data Integrity with Flows in Power Grid Data
arXiv:2603.11745v1 Announce Type: new Abstract: Real-world multivariate time series, particularly in critical infrastructure such as electrical power grids, are often corrupted by noise and anomalies that degrade the performance of downstream tasks. Standard data cleaning approaches often rely on disjoint...
TimeSqueeze: Dynamic Patching for Efficient Time Series Forecasting
arXiv:2603.11352v1 Announce Type: new Abstract: Transformer-based time series foundation models face a fundamental trade-off in choice of tokenization: point-wise embeddings preserve temporal fidelity but scale poorly with sequence length, whereas fixed-length patching improves efficiency by imposing uniform boundaries that may...
Evaluating Explainable AI Attribution Methods in Neural Machine Translation via Attention-Guided Knowledge Distillation
arXiv:2603.11342v1 Announce Type: new Abstract: The study of the attribution of input features to the output of neural network models is an active area of research. While numerous Explainable AI (XAI) techniques have been proposed to interpret these models, the...
Artificial Intelligence for Sentiment Analysis of Persian Poetry
arXiv:2603.11254v1 Announce Type: new Abstract: Recent advancements of the Artificial Intelligence (AI) have led to the development of large language models (LLMs) that are capable of understanding, analysing, and creating textual data. These language models open a significant opportunity in...
Markovian Generation Chains in Large Language Models
arXiv:2603.11228v1 Announce Type: new Abstract: The widespread use of large language models (LLMs) raises an important question: how do texts evolve when they are repeatedly processed by LLMs? In this paper, we define this iterative inference process as Markovian generation...
From Debate to Deliberation: Structured Collective Reasoning with Typed Epistemic Acts
arXiv:2603.11781v1 Announce Type: new Abstract: Multi-agent LLM systems increasingly tackle complex reasoning, yet their interaction patterns remain limited to voting, unstructured debate, or pipeline orchestration. None model deliberation: a phased process where differentiated participants exchange typed reasoning moves, preserve disagreements,...
Governing Evolving Memory in LLM Agents: Risks, Mechanisms, and the Stability and Safety Governed Memory (SSGM) Framework
arXiv:2603.11768v1 Announce Type: new Abstract: Long-term memory has emerged as a foundational component of autonomous Large Language Model (LLM) agents, enabling continuous adaptation, lifelong multimodal learning, and sophisticated reasoning. However, as memory systems transition from static retrieval databases to dynamic,...
Understanding Wikidata Qualifiers: An Analysis and Taxonomy
arXiv:2603.11767v1 Announce Type: new Abstract: This paper presents an in-depth analysis of Wikidata qualifiers, focusing on their semantics and actual usage, with the aim of developing a taxonomy that addresses the challenges of selecting appropriate qualifiers, querying the graph, and...
Where Matters More Than What: Decoding-aligned KV Cache Compression via Position-aware Pseudo Queries
arXiv:2603.11564v1 Announce Type: new Abstract: The Key-Value (KV) cache is crucial for efficient Large Language Models (LLMs) inference, but excessively long contexts drastically increase KV cache memory footprint. Existing KV cache compression methods typically rely on input-side attention patterns within...
Streaming Translation and Transcription Through Speech-to-Text Causal Alignment
arXiv:2603.11578v1 Announce Type: new Abstract: Simultaneous machine translation (SiMT) has traditionally relied on offline machine translation models coupled with human-engineered heuristics or learned policies. We propose Hikari, a policy-free, fully end-to-end model that performs simultaneous speech-to-text translation and streaming transcription...
Multi-Task Reinforcement Learning for Enhanced Multimodal LLM-as-a-Judge
arXiv:2603.11665v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) have been widely adopted as MLLM-as-a-Judges due to their strong alignment with human judgment across various visual tasks. However, most existing judge models are optimized for single-task scenarios and struggle...
SemBench: A Universal Semantic Framework for LLM Evaluation
arXiv:2603.11687v1 Announce Type: new Abstract: Recent progress in Natural Language Processing (NLP) has been driven by the emergence of Large Language Models (LLMs), which exhibit remarkable generative and reasoning capabilities. However, despite their success, evaluating the true semantic understanding of...
Semi-Synthetic Parallel Data for Translation Quality Estimation: A Case Study of Dataset Building for an Under-Resourced Language Pair
arXiv:2603.11743v1 Announce Type: new Abstract: Quality estimation (QE) plays a crucial role in machine translation (MT) workflows, as it serves to evaluate generated outputs that have no reference translations and to determine whether human post-editing or full retranslation is necessary....
Compression Favors Consistency, Not Truth: When and Why Language Models Prefer Correct Information
arXiv:2603.11749v1 Announce Type: new Abstract: Why do language models sometimes prefer correct statements even when trained on mixed-quality data? We introduce the Compression--Consistency Principle: next-token prediction favors hypotheses that allow shorter and more internally consistent descriptions of the training data....
Legal-DC: Benchmarking Retrieval-Augmented Generation for Legal Documents
arXiv:2603.11772v1 Announce Type: new Abstract: Retrieval-Augmented Generation (RAG) has emerged as a promising technology for legal document consultation, yet its application in Chinese legal scenarios faces two key limitations: existing benchmarks lack specialized support for joint retriever-generator evaluation, and mainstream...
Large Language Models for Biomedical Article Classification
arXiv:2603.11780v1 Announce Type: new Abstract: This work presents a systematic and in-depth investigation of the utility of large language models as text classifiers for biomedical article classification. The study uses several small and mid-size open source models, as well as...
DatedGPT: Preventing Lookahead Bias in Large Language Models with Time-Aware Pretraining
arXiv:2603.11838v1 Announce Type: new Abstract: In financial backtesting, large language models pretrained on internet-scale data risk introducing lookahead bias that undermines their forecasting validity, as they may have already seen the true outcome during training. To address this, we present...
Bielik-Minitron-7B: Compressing Large Language Models via Structured Pruning and Knowledge Distillation for the Polish Language
arXiv:2603.11881v1 Announce Type: new Abstract: This report details the creation of Bielik-Minitron-7B, a compressed 7.35B parameter version of the Bielik-11B-v3.0 model, specifically optimized for European languages. By leveraging a two-stage compression methodology inspired by the NVIDIA Minitron approach, we combined...
PersonaTrace: Synthesizing Realistic Digital Footprints with LLM Agents
arXiv:2603.11955v1 Announce Type: new Abstract: Digital footprints (records of individuals' interactions with digital systems) are essential for studying behavior, developing personalized applications, and training machine learning models. However, research in this area is often hindered by the scarcity of diverse...
BTZSC: A Benchmark for Zero-Shot Text Classification Across Cross-Encoders, Embedding Models, Rerankers and LLMs
arXiv:2603.11991v1 Announce Type: new Abstract: Zero-shot text classification (ZSC) offers the promise of eliminating costly task-specific annotation by matching texts directly to human-readable label descriptions. While early approaches have predominantly relied on cross-encoder models fine-tuned for natural language inference (NLI),...
To Words and Beyond: Probing Large Language Models for Sentence-Level Psycholinguistic Norms of Memorability and Reading Times
arXiv:2603.12105v1 Announce Type: new Abstract: Large Language Models (LLMs) have recently been shown to produce estimates of psycholinguistic norms, such as valence, arousal, or concreteness, for words and multiword expressions, that correlate with human judgments. These estimates are obtained by...
Cross-Context Review: Improving LLM Output Quality by Separating Production and Review Sessions
arXiv:2603.12123v1 Announce Type: new Abstract: Large language models struggle to catch errors in their own outputs when the review happens in the same session that produced them. This paper introduces Cross-Context Review (CCR), a straightforward method where the review is...
Long-Context Encoder Models for Polish Language Understanding
arXiv:2603.12191v1 Announce Type: new Abstract: While decoder-only Large Language Models (LLMs) have recently dominated the NLP landscape, encoder-only architectures remain a cost-effective and parameter-efficient standard for discriminative tasks. However, classic encoders like BERT are limited by a short context window,...