{"total":247,"items":[{"citing_arxiv_id":"2605.13743","ref_index":24,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"GHGbench: A Unified Multi-Entity, Multi-Task Benchmark for Carbon Emission Prediction","primary_cat":"cs.LG","submitted_at":"2026-05-13T16:20:49+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"GHGbench is a new multi-entity benchmark for company- and building-level carbon emission prediction that shows building tasks are harder, out-of-distribution gaps dominate, and multimodal data aids generalization.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.13595","ref_index":14,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Inducing Artificial Uncertainty in Language Models","primary_cat":"cs.CL","submitted_at":"2026-05-13T14:30:39+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Inducing artificial uncertainty on trivial tasks allows training probes that achieve higher calibration on hard data than standard approaches while retaining performance on easy data.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.13429","ref_index":63,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"TokAlign++: Advancing Vocabulary Adaptation via Better Token Alignment","primary_cat":"cs.CL","submitted_at":"2026-05-13T12:23:24+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"TokAlign++ learns token alignments between LLM vocabularies from monolingual representations to enable faster adaptation, better text compression, and effective token-level distillation across 15 languages with minimal steps.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.13080","ref_index":149,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Learning to See What You Need: Gaze Attention for Multimodal Large Language Models","primary_cat":"cs.CV","submitted_at":"2026-05-13T06:54:09+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Gaze Attention groups visual embeddings into selectable regions and dynamically restricts attention to task-relevant ones, matching dense baselines with up to 90% fewer visual KV entries via added context tokens.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.12991","ref_index":21,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Not Just RLHF: Why Alignment Alone Won't Fix Multi-Agent Sycophancy","primary_cat":"cs.LG","submitted_at":"2026-05-13T04:45:08+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Pretrained base models exhibit higher yield to peer disagreement than RLHF instruct variants, with the effect localized to mid-layer attention and mitigated by structured dissent rather than prompt defenses.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.12922","ref_index":27,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"When Attention Closes: How LLMs Lose the Thread in Multi-Turn Interaction","primary_cat":"cs.AI","submitted_at":"2026-05-13T02:58:18+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Attention to goal tokens declines in multi-turn LLM interactions while residual representations often retain decodable goal information, and the gap between these predicts whether goal-conditioned behavior survives.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.12756","ref_index":42,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Uncovering Symmetry Transfer in Large Language Models via Layer-Peeled Optimization","primary_cat":"math.OC","submitted_at":"2026-05-12T21:10:34+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Symmetries in next-token prediction targets induce corresponding geometric symmetries such as circulant matrices and equiangular tight frames in the optimal weights and embeddings of a layer-peeled LLM surrogate model.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.12726","ref_index":3,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Before the Last Token: Diagnosing Final-Token Safety Probe Failures","primary_cat":"cs.LG","submitted_at":"2026-05-12T20:30:24+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Final-token probes miss distributed unsafe evidence in jailbreaks, but a PCA-HMM model on prefill trajectories recovers many misses without naive pooling's false positives.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.12446","ref_index":11,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"ORCE: Order-Aware Alignment of Verbalized Confidence in Large Language Models","primary_cat":"cs.LG","submitted_at":"2026-05-12T17:39:43+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"ORCE decouples answer generation from confidence estimation in LLMs and applies rank-based reinforcement learning on sampled completions to better align verbalized confidence with actual correctness likelihood.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.12345","ref_index":18,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Output Composability of QLoRA PEFT Modules for Plug-and-Play Attribute-Controlled Text Generation","primary_cat":"cs.CL","submitted_at":"2026-05-12T16:21:43+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Summing outputs from separately trained QLoRA PEFT modules provides strong performance for attribute-controlled text generation, often matching or exceeding single-task modules even on single-attribute tests.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.12335","ref_index":66,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"EHR-RAGp: Retrieval-Augmented Prototype-Guided Foundation Model for Electronic Health Records","primary_cat":"cs.IR","submitted_at":"2026-05-12T16:17:03+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"EHR-RAGp is a retrieval-augmented EHR foundation model that employs prototype-guided retrieval to dynamically integrate relevant historical patient context, outperforming prior models on clinical prediction tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.12332","ref_index":28,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Towards Automated Air Traffic Safety Assessment Around Non-Towered Airports Using Large Language Models","primary_cat":"cs.AI","submitted_at":"2026-05-12T16:15:15+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Large language models achieve macro F1 scores above 0.85 on binary nominal-versus-danger classification from CTAF radio transcripts and METAR weather data using a new synthetic dataset with a 12-category hazard taxonomy.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.12185","ref_index":28,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Mitigating Context-Memory Conflicts in LLMs through Dynamic Cognitive Reconciliation Decoding","primary_cat":"cs.CL","submitted_at":"2026-05-12T14:29:45+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"DCRD uses attention-map analysis to detect context-memory conflicts in LLMs and conditionally applies either greedy or fidelity-based dynamic decoding, achieving SOTA results on QA tasks across four models and six datasets.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.12129","ref_index":6,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"It's Not the Size: Harness Design Determines Operational Stability in Small Language Models","primary_cat":"cs.SE","submitted_at":"2026-05-12T13:50:30+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"A structured 4-stage pipeline harness raises task success rates to 95%+ in 2-3B parameter models while revealing format collapse and non-monotonic effects when harness support is removed.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.11996","ref_index":43,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"BadSKP: Backdoor Attacks on Knowledge Graph-Enhanced LLMs with Soft Prompts","primary_cat":"cs.AI","submitted_at":"2026-05-12T11:46:25+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"BadSKP poisons graph node embeddings to steer soft prompts in KG-enhanced LLMs, achieving high attack success rates where text-channel backdoors fail due to semantic anchoring.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.11712","ref_index":78,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Toward Stable Value Alignment: Introducing Independent Modules for Consistent Value Guidance","primary_cat":"cs.AI","submitted_at":"2026-05-12T08:02:34+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"SVGT adds independent value modules and Bridge Tokens to LLMs to maintain consistent value guidance, cutting harmful outputs by over 70% in tests while preserving fluency.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.11706","ref_index":36,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"GRAFT: Graph-Tokenized LLMs for Tool Planning","primary_cat":"cs.LG","submitted_at":"2026-05-12T07:59:41+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"GRAFT internalizes tool dependency graphs via dedicated special tokens in LLMs and applies on-policy context distillation to achieve higher exact sequence matching and dependency legality than prior external-graph methods.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.11612","ref_index":34,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"When Emotion Becomes Trigger: Emotion-style dynamic Backdoor Attack Parasitising Large Language Models","primary_cat":"cs.CL","submitted_at":"2026-05-12T06:42:36+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Paraesthesia is an emotion-style dynamic backdoor attack achieving ~99% success rate on instruction and classification tasks across four LLMs while preserving clean performance.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.11592","ref_index":89,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"SoK: Unlearnability and Unlearning for Model Dememorization","primary_cat":"cs.LG","submitted_at":"2026-05-12T06:17:13+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"The first integrated taxonomy, empirical study of interplay and shallow dememorization, plus a theoretical guarantee on dememorization depth for certified unlearning.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"or approximate non-convex losses via mixed linear approxima- tions [58]. However, these methods often depend on smooth losses or positive-definite Hessians, assumptions that may be violated by modern activation functions and loss formulations. To relax these assumptions, recent studies introduce stochastic post-processing and privacy-amplification frameworks for DNN unlearning [89, 124]. These methods combine noisy fine-tuning with tailored gradient operations to enforce DP-style guarantees, avoiding restrictive smoothness assumptions and improving appli- cability to large-scale non-convex models. Further advances reduce the cost of Hessian estimation by attaching unlearning statistics to training samples and leveraging Hessian-vector products [134]."},{"citing_arxiv_id":"2605.11448","ref_index":10,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Deep Minds and Shallow Probes","primary_cat":"cs.LG","submitted_at":"2026-05-12T02:59:44+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Symmetry under affine reparameterizations of hidden coordinates selects a unique hierarchy of shallow coordinate-stable probes and a probe-visible quotient for cross-model transfer.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.10831","ref_index":6,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"SLIM: Sparse Latent Steering for Interpretable and Property-Directed LLM-Based Molecular Editing","primary_cat":"cs.LG","submitted_at":"2026-05-11T16:47:25+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"SLIM decomposes LLM hidden states via sparse autoencoders with learnable gates to enable precise, interpretable steering of molecular properties, yielding up to 42.4-point gains on the MolEditRL benchmark.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.10825","ref_index":14,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Large Spectrum Models (LSMs): Decoder-Only Transformer-Powered Spectrum Activity Forecasting via Tokenized RF Data","primary_cat":"cs.NI","submitted_at":"2026-05-11T16:43:55+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Decoder-only transformers trained on tokenized RF spectrum data from 22 TB of measurements achieve 3.25 dB RMSE in spectrum activity forecasting across 33 bands.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.10772","ref_index":12,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Towards a Large Language-Vision Question Answering Model for MSTAR Automatic Target Recognition","primary_cat":"cs.CV","submitted_at":"2026-05-11T16:05:58+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"A fine-tuned large language-vision model achieves 98% accuracy on visual question answering for military vehicle identification in SAR imagery from an extended MSTAR benchmark.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.10639","ref_index":12,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Navigating the Sea of LLM Evaluation: Investigating Bias in Toxicity Benchmarks","primary_cat":"cs.AI","submitted_at":"2026-05-11T14:27:39+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Toxicity benchmarks for LLMs produce inconsistent results when task type, input domain, or model changes, revealing intrinsic evaluation biases.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.10606","ref_index":47,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Measuring Embedding Sensitivity to Authorial Style in French: Comparing Literary Texts with Language Model Rewritings","primary_cat":"cs.CL","submitted_at":"2026-05-11T14:05:46+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Embeddings reliably capture authorial stylistic features in French literary texts, and these signals persist after LLM rewriting while showing model-specific patterns.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.10043","ref_index":23,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Personalizing LLMs with Binary Feedback: A Preference-Corrected Optimization Framework","primary_cat":"cs.CL","submitted_at":"2026-05-11T06:12:24+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"C-BPO personalizes LLMs via preference-calibrated binary signals and PU learning theory to isolate inter-user differences from shared task knowledge.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.09875","ref_index":21,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Cross-Family Universality of Behavioral Axes via Anchor-Projected Representations","primary_cat":"cs.AI","submitted_at":"2026-05-11T02:01:18+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Behavioral directions from one LLM family transfer to others via projection into a shared anchor coordinate space, yielding 0.83 ten-way detection accuracy and steering effects up to 0.46% on held-out models.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.09808","ref_index":62,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Quantifying the Utility of User Simulators for Building Collaborative LLM Assistants","primary_cat":"cs.CL","submitted_at":"2026-05-10T23:06:24+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Fine-tuned simulators grounded in real human data produce LLM assistants that win more often against real users than those trained against role-playing simulators.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.09745","ref_index":2,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Entropy-informed Decoding: Adaptive Information-Driven Branching","primary_cat":"cs.LG","submitted_at":"2026-05-10T20:44:47+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"EDEN adaptively sets branching factor proportional to next-token entropy, achieving better accuracy per expansion than fixed beam search while providing a proof that monotone entropy-based branching outperforms any fixed budget allocation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.09665","ref_index":10,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Learning Multi-Indicator Weights for Data Selection: A Joint Task-Model Adaptation Framework with Efficient Proxies","primary_cat":"cs.LG","submitted_at":"2026-05-10T17:30:16+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"A joint task-model adaptation method learns optimal weights for data selection indicators via ICL proxies on small validation sets, matching or exceeding full-dataset fine-tuning performance with only 30% of samples on GSM8K.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.09472","ref_index":49,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Positional LSH: Binary Block Matrix Approximation for Attention with Linear Biases","primary_cat":"cs.LG","submitted_at":"2026-05-10T10:58:20+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"ALiBi bias is the expectation of positional LSH-induced block masks, yielding spectral and max-norm approximation bounds that reduce long-context biased attention to randomized short-context unbiased attention.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.09391","ref_index":3,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Do Linear Probes Generalize Better in Persona Coordinates?","primary_cat":"cs.AI","submitted_at":"2026-05-10T07:38:34+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Probes on persona principal components from contrastive prompts generalize better than raw activation probes for harmful behaviors across 10 datasets.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.09176","ref_index":14,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Navigating LLM Valley: From AdamW to Memory-Efficient and Matrix-Based Optimizers","primary_cat":"cs.LG","submitted_at":"2026-05-09T21:34:28+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"This survey organizes LLM optimizer literature into categories and argues the field is shifting toward rigorous, multi-factor comparisons of convergence, memory, stability, and complexity.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.08964","ref_index":84,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Trustworthy AI: Ensuring Reliability and Accountability from Models to Agents","primary_cat":"cs.LG","submitted_at":"2026-05-09T14:11:36+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"The thesis presents a kernel method for multiaccuracy across overlooked subpopulations, information-theoretic optimal watermarking for LLMs, and a simulator showing LLM agents outperforming humans in supply chains while creating tail risks.","context_count":1,"top_context_role":"dataset","top_context_polarity":"use_dataset","context_text":"We evaluate the detection-distortion-robustness tradeoff of HeavyWater and SimplexWater following the experimental benchmarking guidelines in Waterbench [153] and MarkMyWords [126]. We use two popular prompt-generation datasets: Finance- QA [110] (covering Q&A tasks) and LCC [ 25] (covering coding tasks). We evaluate on three open-weight models: Llama2-7B [ 151], Llama3-8B [64], and Mistral-7B [ 84]. Implementation details are provided in Appendix C.5.4, and full results, including ablation studies, are presented in Appendix C.6. Baseline Watermarks.We benchmark SimplexWater and HeavyWater against the Gumbel watermark [ 2], the Inverse-transform watermark [ 93], the Correlated Channel watermark [108], and the SynthID watermark [37] with binary scores and K = 15 competition"},{"citing_arxiv_id":"2605.08898","ref_index":72,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"LLM-Agnostic Semantic Representation Attack","primary_cat":"cs.CL","submitted_at":"2026-05-09T11:43:47+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"SRA achieves 99.71% average attack success across 26 LLMs by optimizing for coherent malicious semantics via the SRHS algorithm, with claimed theoretical guarantees on convergence and transfer.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.08863","ref_index":14,"ref_count":2,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Max-pooling Network Revisited: Analyzing the Role of Semantic Probability in Multiple Instance Learning for Hallucination Detection","primary_cat":"cs.CL","submitted_at":"2026-05-09T10:30:30+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"A lightweight max-pooling network with MLP detects LLM hallucinations competitively without semantic consistency computations by adaptively aggregating internal token features.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.08840","ref_index":12,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"ReST-KV: Robust KV Cache Eviction with Layer-wise Output Reconstruction and Spatial-Temporal Smoothing","primary_cat":"cs.CL","submitted_at":"2026-05-09T09:49:32+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"ReST-KV formulates KV eviction as layer-wise output reconstruction optimization with spatial-temporal smoothing, outperforming baselines by 2.58% on LongBench and 15.2% on RULER while cutting decoding latency by 10.61x at 128k context.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.08734","ref_index":15,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"AdaPreLoRA: Adafactor Preconditioned Low-Rank Adaptation","primary_cat":"cs.LG","submitted_at":"2026-05-09T06:37:44+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"AdaPreLoRA pairs the Adafactor diagonal Kronecker preconditioner on the full weight matrix with a closed-form factor-space solve that selects the update minimizing an H_t-weighted imbalance, yielding competitive results on GPT-2, Mistral-7B, Qwen2-7B and diffusion personalization tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.08533","ref_index":35,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Human-LLM Dialogue Improves Diagnostic Accuracy in Emergency Care","primary_cat":"cs.AI","submitted_at":"2026-05-08T22:40:10+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Interactive LLM dialogue raised residents' hard-case diagnostic correctness from 0.589 to 0.734 and produced medium effect sizes in a blinded study of seven physicians on 52 emergency cases.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.08378","ref_index":144,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Reinforcement Learning for Scalable and Trustworthy Intelligent Systems","primary_cat":"cs.LG","submitted_at":"2026-05-08T18:36:25+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"Reinforcement learning is advanced for communication-efficient federated optimization and for preference-aligned, contextually safe policies in large language models.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.07925","ref_index":2,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"How Value Induction Reshapes LLM Behaviour","primary_cat":"cs.CL","submitted_at":"2026-05-08T15:58:42+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Inducing targeted values in LLMs through fine-tuning causes spillover to related or opposing values, boosts safety metrics, and increases anthropomorphic and sycophantic language across all tested values.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.07234","ref_index":23,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Reformulating KV Cache Eviction Problem for Long-Context LLM Inference","primary_cat":"cs.CL","submitted_at":"2026-05-08T04:37:22+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"LaProx reformulates KV cache eviction as an output-aware matrix approximation, enabling a unified global token selection strategy that preserves LLM performance at 5% cache size across long-context benchmarks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.07138","ref_index":6,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Can You Break RLVER? Probing Adversarial Robustness of RL-Trained Empathetic Agents","primary_cat":"cs.AI","submitted_at":"2026-05-08T02:14:12+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"RLVER agents improve emotional responsiveness under adversarial user behaviors but exhibit no measurable gains in tracking emotional states compared to untuned base models.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.07105","ref_index":29,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Theoretical Limits of Language Model Alignment","primary_cat":"cs.LG","submitted_at":"2026-05-08T01:32:22+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"The maximum reward gain under KL-regularized LM alignment is a Jeffreys divergence term, estimable as covariance from base samples, with best-of-N approaching the theoretical limit.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.07096","ref_index":33,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Query-efficient model evaluation using cached responses","primary_cat":"cs.LG","submitted_at":"2026-05-08T01:24:06+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"DKPS-based methods leverage cached model responses to achieve equivalent benchmark prediction accuracy with substantially fewer queries than standard evaluation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.07075","ref_index":50,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"ModelLens: Finding the Best for Your Task from Myriads of Models","primary_cat":"cs.LG","submitted_at":"2026-05-08T00:49:05+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"ModelLens learns a performance-aware latent space from 1.62M leaderboard records to rank unseen models on unseen datasets without forward passes on the target.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.06765","ref_index":37,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"VITA-QinYu: Expressive Spoken Language Model for Role-Playing and Singing","primary_cat":"cs.CL","submitted_at":"2026-05-07T17:59:56+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"VITA-QinYu is the first expressive end-to-end spoken language model supporting role-playing and singing alongside conversation, trained on 15.8K hours of data and outperforming prior models on expressiveness and conversational benchmarks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.06632","ref_index":33,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Crafting Reversible SFT Behaviors in Large Language Models","primary_cat":"cs.LG","submitted_at":"2026-05-07T17:44:07+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":8.0,"formal_verification":"none","one_line_summary":"LCDD creates sparse carriers for SFT behaviors that SFT-Eraser can reverse, with ablations showing the sparse structure enables causal control.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.06476","ref_index":3,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Towards Emotion Consistency Analysis of Large Language Models in Emotional Conversational Contexts","primary_cat":"cs.CL","submitted_at":"2026-05-07T16:01:48+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"LLMs show below-average consistency and vulnerability to false beliefs in emotional queries with false presuppositions, more so for moderate emotions.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.06221","ref_index":12,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"UniPrefill: Universal Long-Context Prefill Acceleration via Block-wise Dynamic Sparsification","primary_cat":"cs.CL","submitted_at":"2026-05-07T13:18:08+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"UniPrefill accelerates LLM prefill via block-wise dynamic sparsification, achieving up to 2.1x TTFT speedup while supporting hybrid architectures and native vLLM continuous batching.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}