{"total":15,"items":[{"citing_arxiv_id":"2607.02510","ref_index":23,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Online Safety Monitoring for LLMs","primary_cat":"cs.AI","submitted_at":"2026-07-02T17:59:43+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"Simple thresholding on an external verifier signal, calibrated by risk control, performs competitively with sequential hypothesis testing monitors on math reasoning and red-teaming datasets.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2607.02206","ref_index":11,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Prediction Sets for Counterfactual Decisions: Coverage, Optimality, and Conformal Prediction","primary_cat":"stat.ML","submitted_at":"2026-07-02T14:13:08+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Introduces policy-coupled coverage for conformal prediction in counterfactual decisions and the PC-RACP procedure that achieves higher utility with finite-sample coverage guarantees.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.29654","ref_index":34,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Budgeted Act-or-Defer Multi-Agent LLM Deliberation with Local Reliability Bounds","primary_cat":"cs.AI","submitted_at":"2026-06-28T23:46:56+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A kNN lower-confidence-bound approach for act-or-defer decisions in multi-agent LLM debates respects user-declared wrong-action budgets while achieving high automation rates on benchmarks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.12587","ref_index":66,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Strategic Decision Support for AI Agents","primary_cat":"cs.AI","submitted_at":"2026-06-10T18:34:36+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"The paper introduces an optimization framework for AI agents to strategically seek support, proving a threshold policy on support value and providing an online algorithm to control missed-support error without distributional assumptions.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.08277","ref_index":38,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Remember with Confidence: Uncertainty Quantification for Spatio-temporal Memory with Probabilistic Guarantees","primary_cat":"cs.CV","submitted_at":"2026-06-06T17:55:27+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Introduces object-level semantic uncertainty for VLM memory, the UQ-DAAAM refinement system, and probabilistic guarantees that selected high-quality views reduce uncertainty more effectively.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.00251","ref_index":50,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Capability Self-Assessment: Teaching LLMs to Know Their Limits","primary_cat":"cs.AI","submitted_at":"2026-05-29T18:32:14+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Reinforcement learning teaches LLMs to assess their own capabilities more effectively than supervised fine-tuning, preserves original skills, generalizes out of distribution, and aids local-cloud routing and data selection.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.28264","ref_index":38,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Entropy Distribution as a Fingerprint for Hallucinations in Generative Models","primary_cat":"cs.AI","submitted_at":"2026-05-27T10:12:44+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Token entropy distributions fingerprint hallucinations in generative models, enabling the Calibrated Entropy Score (CES) for single-pass black-box detection with calibration guarantees via a novel DKW inequality.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.23189","ref_index":28,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Empirical Bayes Conformal Prediction for Vision and Language Models","primary_cat":"cs.LG","submitted_at":"2026-05-22T03:17:14+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Empirical Bayes conformal prediction converts score variability into r-value nonconformity scores that preserve target coverage while reducing inclusion of high-variance false candidates in image classification, CLIP VLMs, and LLMs.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.16407","ref_index":49,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Proof-Carrying Certificates for LLM Pipelines: A Trust-Boundary Architecture","primary_cat":"cs.LO","submitted_at":"2026-05-13T12:01:41+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"partial","one_line_summary":"Introduces a trust-boundary architecture in Lean 4 with three certificate families and two operators that deliver sorry-free, axiom-audited assurances for LLM pipeline components.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.27914","ref_index":28,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Geometry-Calibrated Conformal Abstention for Language Models","primary_cat":"cs.CL","submitted_at":"2026-04-30T14:20:16+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Geometry-calibrated conformal abstention lets language models abstain from uncertain queries with finite-sample guarantees on both participation rate and conditional correctness of answers.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.17487","ref_index":12,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Answer Only as Precisely as Justified: Calibrated Claim-Level Specificity Control for Agentic Systems","primary_cat":"cs.CL","submitted_at":"2026-04-19T15:20:02+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Compositional selective specificity (CSS) decomposes generated answers into claims and emits each at the most specific level supported by evidence, raising overcommitment-aware utility from 0.846 to 0.913 on LongFact while retaining 0.938 specificity.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.01413","ref_index":17,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Adaptive Stopping for Multi-Turn LLM Reasoning","primary_cat":"cs.CL","submitted_at":"2026-04-01T21:22:19+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":8.0,"formal_verification":"none","one_line_summary":"MiCP is the first conformal prediction method for multi-turn LLM pipelines that allocates per-turn error budgets to enable adaptive stopping with an overall coverage guarantee, shown to reduce turns and cost on RAG and ReAct benchmarks.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"Stage 2: Prediction Set and Early Stopping.Let {(xi, yi, {Ai t}T t=0)}ncal i=1 be the calibration set, where yi is the gold answer and Ai t is the set of M sampled answers at turn t using filtered context from Stage 1. We jointly learn: (1) turn-specific stopping thresholds { ˆqt}T t=0 that determine whether the model should stop after turnt, and (2) a final prediction set C(x i)⊆ T[ t=0 Ai t ∪ {Can't Answer}. (7) At each turn t, MiCP computes a calibrated confidence score from the sampled answers and stops whenever the score exceeds ˆqt. If the model stops at turn t<T , the prediction 5 Preprint. Under review. set is constructed fromSt τ=0 Ai τ; after the final turn T, it may additionally include \"Can't Answer\". The prediction set is calibrated to satisfy P yi ∈ C(x i)ify i ∈ T[ t=0 Ai t,Can't Answer∈ C(x i)ify i /∈ T[ t=0 Ai t ! ≥1−α. (8) That is, MiCP guarantees the prediction set either contains the correct answer when it appears among the sampled answers, or abstains via \"Can't Answer\" when no sampled answer is correct. 4.2 Retrieval Threshold Calibration A key challenge in multi-turn pipelines is that each retrieval turn returns a fixed set of top-K passages, many of which may be irrelevant. As retrieval proceeds, these irrelevant passages accumulate and pollute the LLM's context, degrading both answer quality and stopping decisions. Prior work has applied CP to filter irrelevant passages in single-round retrieval (Chakraborty et al., 2025); we extend this to the multi-turn setting, applying retrieval filtering at every turn while preserving coverage over the entire reasoning process. Because a gold passage p may be retrieved at multiple turns with different relevance scores, we define its optimistic relevance score as the maximum across all turns: s∗(p,x i) =max t:p∈Top-K(r i t) st(p). (9) We aggregate these scores across the calibration set: Sret ={s ∗(p, xi)|p∈ P ∗ i , i= 1, . . . ,ncal}, and define the conformal retrieval threshold as ˆqret =Quan"},{"citing_arxiv_id":"2602.07633","ref_index":35,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Flow-Based Conformal Predictive Distributions","primary_cat":"stat.ML","submitted_at":"2026-02-07T17:26:50+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Differentiable nonconformity scores induce flows that sample conformal prediction set boundaries, and mixing flows across levels produces conformal predictive distributions whose quantiles match the sets.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2510.04406","ref_index":21,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Decomposition-Based Modular Conformal Prediction for Two-Stage Modeling","primary_cat":"stat.ML","submitted_at":"2025-10-06T00:33:18+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"A decomposition-based modular conformal prediction method for two-stage models with FWER-controlled stage-wise scaling and adaptive extension for non-stationary data.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2308.05374","ref_index":117,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Trustworthy LLMs: a Survey and Guideline for Evaluating Large Language Models' Alignment","primary_cat":"cs.AI","submitted_at":"2023-08-10T06:43:44+00:00","verdict":"ACCEPT","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Survey organizes LLM trustworthiness into seven categories and 29 sub-categories, measures eight sub-categories on popular models, and finds that more aligned models generally score higher but with varying effectiveness.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"[115] Adam Fisch, Tal Schuster, Tommi Jaakkola, and Regina Barzilay. Efficient conformal prediction via cascaded inference with expanded admission. arXiv preprint arXiv:2007.03114, 2020. [116] Bhawesh Kumar, Charlie Lu, Gauri Gupta, Anil Palepu, David Bellamy, Ramesh Raskar, and Andrew Beam. Conformal prediction with large language models for multi-choice question answering. arXiv preprint arXiv:2305.18404, 2023. [117] Victor Quach, Adam Fisch, Tal Schuster, Adam Yala, Jae Ho Sohn, Tommi S Jaakkola, and Regina Barzilay. Conformal language modeling. arXiv preprint arXiv:2306.10193, 2023. [118] Wenxuan Zhou, Fangyu Liu, and Muhao Chen. Contrastive out-of-distribution detection for pretrained trans- formers. arXiv preprint arXiv:2104.08812, 2021. [119] Dan Hendrycks, Xiaoyuan Liu, Eric Wallace, Adam Dziedzic, Rishabh Krishnan, and Dawn Song."}],"limit":50,"offset":0}