Creates LoCoMo benchmark dataset for very long-term LLM conversational memory and shows current models struggle with lengthy dialogues and long-range temporal dynamics.
hub
Proceedings of the 2023
47 Pith papers cite this work, alongside 220 external citations. Polarity classification is still indexing.
47
Pith papers citing it
220
external citations · Crossref
hub tools
citation-role summary
background 3
citation-polarity summary
roles
background 3polarities
background 3representative citing papers
MedHal-Loc benchmark shows KG-triple hallucination detectors localize errors no better than chance on controlled medical statements due to entity extraction limits, while NLI and consistency methods succeed above chance, and real hallucinations are mostly diffuse conclusion changes.
VOIR DIRE benchmark shows MLLM-as-a-Judge systems decompose into positivity-floor calibration failure and orientation failure on culturally contested items, with persona prompting recovering only the former.
AuthorityBench shows citation presence (real or fabricated) increases LLM hallucination rates vs no-citation baseline, strongest for fabricated citations on true claims, with domain variation but negligible venue or author effects.
WorldReasoner supplies 345 resolved forecasting tasks built from 14,141 articles to score LM agents on outcome quality, evidence quality, and reasoning quality against time-bounded evidence and hindsight graphs.
A new benchmark and clean-room harness show frontier AI agents reach only 0.337 factual F1 when synthesizing conclusions from scientific evidence.
PhantomBench is a new benchmark of 60K+ non-existent terms showing language models hallucinate at rates up to 86.7 percent even when inputs assume the concepts exist.
Commercial AI chatbots reach over 90% multiple-choice accuracy on recent news facts but lose 11-17% in free response and drop to 19-70% on subtle false-premise questions, with retrieval failures causing most errors and clear Anglophone bias.
MemDocAgent generates consistent hierarchical repository-level code documentation by combining dependency-aware traversal with memory-guided agent interactions that accumulate work traces.
LLMs routinely produce unsupported causal stories for personal sensing anomalies, and richer evidence or constrained prompts do not reliably eliminate this epistemic overreach.
Transpose-invariant spectral diagnostics on attention operators are orientation-blind, and a φ-G two-axis diagnostic distinguishes hallucination modes with 0.62-0.84 LC-AUROC and predicted polarity reversal.
SemGrad measures LLM uncertainty via gradients in semantic space using a Semantic Preservation Score to select embeddings, with HybridGrad combining it with parameter gradients to outperform sampling-based baselines especially when multiple responses are valid.
SOB benchmark shows LLMs achieve near-perfect schema compliance but value accuracy of only 83% on text, 67% on images, and 24% on audio.
Chain-of-illocution prompting improves source adherence in RAG explanations for programming education by up to 63% over baselines.
TSVer is a new benchmark dataset for fact verification against time-series evidence, with 304 annotated real-world claims, 400 time series, verdicts, and justifications, plus baseline results showing current models struggle.
Hallucinations arise from biased latent inference paths rather than missing knowledge, demonstrated via a new diagnostic testbed TrapQA that isolates task-retrieval and key-selection biases.
LoFa is a new benchmark and LFR@k metric for measuring LLM resistance to sustained logical fallacy attacks via generated question-argument pairs and debate simulations.
Global calibration metrics like ECE are confounded by accuracy; the proposed ACE framework with three accuracy-controlled views shows many prior calibration advantages weaken or reverse.
Introduces claim-conditioned re-scoring (SIFT) and warranted supports proportion (WSP) metric, reporting accuracy recovery up to 27.6 points and WSP calibration at AUC 0.92 on FEVER, SciFact and other benchmarks.
MÖVE presents a new German-language benchmark evaluating 39 LLMs on performance and governance criteria using ten public-administration datasets.
GIScholarBench shows LLMs exhibit consistent overconfidence across three scholarly tasks in GIS, with different manifestations in factual retrieval, citation expansion, and idea generation.
A new evaluation framework shows that even the best tested LLM only reliably adjusts response complexity in the intended direction 46% of the time across 98 scientific queries.
RISC reformulates self-consistency answer selection as a ranking task solved by a lightweight LambdaRank model with five hand-designed features, yielding better accuracy-efficiency trade-offs than majority voting on QA benchmarks.
Skill-RM unifies heterogeneous reward criteria by modeling reward computation as dynamic execution of a reusable Reward-Evaluation Skill within an agent framework.
citing papers explorer
-
Evaluating Very Long-Term Conversational Memory of LLM Agents
Creates LoCoMo benchmark dataset for very long-term LLM conversational memory and shows current models struggle with lengthy dialogues and long-range temporal dynamics.
-
MedHal-Loc: Are "Explainable-by-Architecture" Medical Hallucination Detectors Faithful Localizers? A Localization Benchmark
MedHal-Loc benchmark shows KG-triple hallucination detectors localize errors no better than chance on controlled medical statements due to entity extraction limits, while NLI and consistency methods succeed above chance, and real hallucinations are mostly diffuse conclusion changes.
-
Jury Duty: Calibration and Orientation Failures in MLLM-as-a-Judge Under Cultural Ambiguity
VOIR DIRE benchmark shows MLLM-as-a-Judge systems decompose into positivity-floor calibration failure and orientation failure on culturally contested items, with persona prompting recovering only the former.
-
Authority, Truth, and Citation Bias: A Large-Scale Multi-Domain Benchmark for Studying Epistemic Susceptibility in Large Language Models
AuthorityBench shows citation presence (real or fabricated) increases LLM hallucination rates vs no-citation baseline, strongest for fabricated citations on true claims, with domain variation but negligible venue or author effects.
-
WorldReasoner: Evaluating Whether Language Model Agents Forecast Events with Valid Reasoning
WorldReasoner supplies 345 resolved forecasting tasks built from 14,141 articles to score LM agents on outcome quality, evidence quality, and reasoning quality against time-bounded evidence and hindsight graphs.
-
Can AI Agents Synthesize Scientific Conclusions?
A new benchmark and clean-room harness show frontier AI agents reach only 0.337 factual F1 when synthesizing conclusions from scientific evidence.
-
PhantomBench: Benchmarking the Non-existential Threat of Language Models
PhantomBench is a new benchmark of 60K+ non-existent terms showing language models hallucinate at rates up to 86.7 percent even when inputs assume the concepts exist.
-
Remember Your Trace: Memory-Guided Long-Horizon Agentic Framework for Consistent and Hierarchical Repository-Level Code Documentation
MemDocAgent generates consistent hierarchical repository-level code documentation by combining dependency-aware traversal with memory-guided agent interactions that accumulate work traces.
-
Self-Attention as Transport: Limits of Symmetric Spectral Diagnostics
Transpose-invariant spectral diagnostics on attention operators are orientation-blind, and a φ-G two-axis diagnostic distinguishes hallucination modes with 0.62-0.84 LC-AUROC and predicted polarity reversal.
-
Gradients with Respect to Semantics Preserving Embeddings Tell the Uncertainty of Large Language Models
SemGrad measures LLM uncertainty via gradients in semantic space using a Semantic Preservation Score to select embeddings, with HybridGrad combining it with parameter gradients to outperform sampling-based baselines especially when multiple responses are valid.
-
TSVer: A Benchmark for Fact Verification Against Time-Series Evidence
TSVer is a new benchmark dataset for fact verification against time-series evidence, with 304 annotated real-world claims, 400 time series, verdicts, and justifications, plus baseline results showing current models struggle.
-
Understanding Why Language Models Hallucinate: Testing Reasoning Against Priors
Hallucinations arise from biased latent inference paths rather than missing knowledge, demonstrated via a new diagnostic testbed TrapQA that isolates task-retrieval and key-selection biases.
-
Truth or Sophistry? LoFa: A Benchmark for LLM Robustness Against Logical Fallacies
LoFa is a new benchmark and LFR@k metric for measuring LLM resistance to sustained logical fallacy attacks via generated question-argument pairs and debate simulations.
-
When Calibration Rankings Reverse: Accuracy-Controlled Evaluation for Fair Comparison of LLMs
Global calibration metrics like ECE are confounded by accuracy; the proposed ACE framework with three accuracy-controlled views shows many prior calibration advantages weaken or reverse.
-
The Warrant Gap: Claim-Conditioned Re-scoring for Fact-Checking
Introduces claim-conditioned re-scoring (SIFT) and warranted supports proportion (WSP) metric, reporting accuracy recovery up to 27.6 points and WSP calibration at AUC 0.92 on FEVER, SciFact and other benchmarks.
-
M\"OVE: A Holistic LLM Benchmark for the German Public Sector
MÖVE presents a new German-language benchmark evaluating 39 LLMs on performance and governance criteria using ten public-administration datasets.
-
GIScholarBench: Benchmarking LLM Overconfidence in GIS Research
GIScholarBench shows LLMs exhibit consistent overconfidence across three scholarly tasks in GIS, with different manifestations in factual retrieval, citation expansion, and idea generation.
-
Explain Like I'm 5 or Whatever I Choose: Evaluating the Interactive Potential of Language Model Responses
A new evaluation framework shows that even the best tested LLM only reliably adjusts response complexity in the intended direction 46% of the time across 98 scientific queries.
-
Boosting Self-Consistency with Ranking
RISC reformulates self-consistency answer selection as a ranking task solved by a lightweight LambdaRank model with five hand-designed features, yielding better accuracy-efficiency trade-offs than majority voting on QA benchmarks.
-
Skill-RM: Unifying Heterogeneous Evaluation Criteria via Agent Skill
Skill-RM unifies heterogeneous reward criteria by modeling reward computation as dynamic execution of a reusable Reward-Evaluation Skill within an agent framework.
-
Resonant Context Anchoring: Decoupling Attention Routing and Signal Gain at Inference Time
RCA is a training-free module that boosts input context signal strength in the residual stream of LLMs by orthogonal decoupling of attention routing from value magnitude.
-
Grounded Decoding: Retrieval-Anchored Probability Fusion for Faithful RAG
Grounded Decoding fuses full-RAG and retrieval-only next-token distributions via normalized geometric mean from a KL-barycenter to improve factual consistency and citation quality in RAG.
-
KG-Guard: Graph-Based Hallucination Detection for Knowledge Base Question Answering
KG-Guard augments knowledge graphs with a virtual question node and uses a graph encoder plus MLP to classify LLM-proposed answers as hallucinations or not, reporting superior F1 scores and downstream improvements on three benchmarks.
-
Argus: Evidence Assembly for Scalable Deep Research Agents
Argus coordinates a Navigator and multiple Searchers via an evidence graph for deep research, reporting average gains of 5.5 points with one Searcher and 12.7 points with eight parallel Searchers across eight benchmarks, reaching 86.2 on BrowseComp with 64 Searchers.
-
Proof-Carrying Certificates for LLM Pipelines: A Trust-Boundary Architecture
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.
-
Logical Consistency as a Bridge: Improving LLM Hallucination Detection via Label Constraint Modeling between Responses and Self-Judgments
LaaB improves LLM hallucination detection by mapping self-judgment labels back into neural feature space and using mutual learning under logical consistency constraints between responses and meta-judgments.
-
LLM Ghostbusters: Surgical Hallucination Suppression via Adaptive Unlearning
Adaptive Unlearning suppresses package hallucinations in code-generating LLMs by 81% while preserving benchmark performance, using model-generated data and no human labels.
-
Architecture Determines Observability of Transformers
Architecture and training determine whether transformers retain a readable internal signal that lets activation monitors catch errors missed by output confidence.
-
Whose Story Gets Told? Positionality and Bias in LLM Summaries of Life Narratives
A proposed pipeline shows LLMs introduce detectable race and gender biases when summarizing life narratives, creating potential for representational harm in research.
-
Beyond Precision: Importance-Aware Recall for Factuality Evaluation in Long-Form LLM Generation
An importance-aware recall metric for LLM factuality evaluation reveals models are better at avoiding false claims than covering all relevant facts.
-
ACE-Bench: A Lightweight Benchmark for Evaluating Azure SDK Usage Correctness
ACE-Bench is an execution-free benchmark that scores LLM coding agents on correct Azure SDK usage via deterministic regex checks and reference-based LLM judges derived from official documentation.
-
Textual Bayes: Quantifying Prompt Uncertainty in LLM-Based Systems
Introduces a Bayesian framework viewing LLM prompts as textual parameters and proposes MHLP, a novel MCMC algorithm using LLM proposals, to perform inference and improve accuracy plus uncertainty quantification on benchmarks.
-
LoVeC: Reinforcement Learning for Better Verbalized Confidence in Long-Form Generations
LoVeC uses RL to train LLMs to output verbalized numerical confidence scores for statements in long-form text, achieving better calibration than self-consistency baselines on QA datasets while being 20x faster.
-
Token-Level Density-Based Uncertainty Quantification Methods for Eliciting Truthfulness of Large Language Models
Adapts multi-layer token-level Mahalanobis distance with supervised linear regression to yield improved uncertainty scores for LLM truthfulness tasks.
-
Unconditional Truthfulness: Learning Unconditional Uncertainty of Large Language Models
A regression model using attention features and recurrent uncertainty scores improves selective generation in LLMs over unsupervised and supervised baselines on ten datasets and three models.
-
When Summaries Distort Decisions: Information Fidelity in LLM-Compressed Financial Analysis
LLM-based compression of financial source material can alter downstream investment decisions via decontextualization and model dependency, addressed by an agentic auditing approach that checks multiple compressions against the original.
-
Only Say What You Know: Calibration-Aware Generation for Long-Form Factuality
Exploration-Commitment Decoupling instantiated as Calibration-Aware Generation improves long-form factuality by up to 13% and reduces decoding time by up to 37% on five benchmarks.
-
IUQ: Interrogative Uncertainty Quantification for Long-Form Large Language Model Generation
IUQ quantifies claim-level uncertainty in long-form LLM generation by combining inter-sample consistency and intra-sample faithfulness through an interrogate-then-respond approach and outperforms baselines on two datasets.
-
Constructing Evaluation Datasets for Procedural Reasoning: Balancing Naturalness, Grounding, and Multi-Hop Coverage
Strict generation directly from Task-Method-Knowledge models yields 96.5% grounded and 92.6% usable QA pairs across 23 topics, outperforming transcript-first and TMK-aware alternatives on representational grounding.
-
Grounding Multi-Hop Reasoning in Structural Causal Models via Group Relative Policy Optimization
An SCM-GRPO framework grounds multi-hop reasoning in structural dependency graphs and optimizes chain length via rule-based RL, outperforming baselines on HoVer and EX-FEVER.
-
A Survey on LLM-as-a-Judge
A survey on LLM-as-a-Judge that reviews reliability strategies, proposes evaluation methods, and introduces a novel benchmark for assessing such systems.