DLR creates discrete latent tokens from rendered CoT images via clustering, enabling up to 20x compression and interpretable trajectories that outperform continuous latent baselines on reasoning tasks.
Canonical reference
Large Language Models: The Need for Nuance in Current Debates and a Pragmatic Perspective on Understanding
Canonical reference. 89% of citing Pith papers cite this work as background.
99
Pith papers citing it
Background
89% of classified citations
citation-role summary
background 24
method 2
dataset 1
citation-polarity summary
co-cited works
representative citing papers
A Gaussian information-gain metric in embedding space quantifies semantic progress in dialogues via uncertainty reduction and shows competitive agreement with human judgments on MT-Bench and UltraFeedback.
τ-Rec is a benchmark for agentic recommender systems with verifiable rewards, RTE mechanism, and pass^k metrics that shows top models reach only ~57% at pass^1 and ~35% at pass^4.
CultureForest benchmark shows top LLMs degrade sharply on open-ended cultural reasoning tasks, exhibit regional disparities, and are limited more by effective use of knowledge than by lack of knowledge itself.
PolySpeech-100 is a new benchmark for native-level speech comprehension across 110 linguistic variants that evaluates 22 models and reports E2E advantages on dialects, robustness gaps on low-resource languages, and degradation from Chain-of-Thought prompting.
Prover-verifier deliberation yields a high-confidence subset of LLM answers with ~30pp higher precision than the complement on GPQA Diamond by using defender-challenger dialogues.
A test-time zeroth-order optimization of prompt embeddings using a bounded self-supervised proxy from demonstration log-probabilities improves ICL accuracy and correlates with gains across tasks.
The paper presents ChildAgentEval as the first psychometrically grounded benchmark comparing MLLM-based agents' reasoning performance to age-specific human cognitive stages.
A cross-modal alignment attack achieves AUC 0.821 for single-sample black-box membership inference on VLMs such as LLaVA-1.5 by quantifying image-generated caption similarity.
LGMT is a logic-grounded metamorphic testing framework that detects hidden reasoning defects in LLMs by checking consistency on semantically invariant inputs derived from FOL equivalences.
StereoTales shows that all tested LLMs emit harmful stereotypes in open-ended stories, with associations adapting to prompt language and targeting locally salient groups rather than transferring uniformly across languages.
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.
COPYCOP identifies copycat GNNs by matching their node embeddings despite architectural differences and adversarial transformations, backed by theoretical guarantees and tests on 14 datasets across 5 architectures.
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.
Chain of Evidence introduces a retriever-agnostic visual attribution method for iRAG that reasons over document screenshots with VLMs to output precise bounding boxes, outperforming text baselines on Wiki-CoE and SlideVQA.
ReformIR adaptively prioritizes reformulations and documents with a surrogate model guided by ranker feedback to boost recall while suppressing drift under fixed reranking budgets.
AI Overviews and Gemini retrieve substantially different sources than traditional Google search (Jaccard similarity <0.2), favor Google-owned content, appear for 51.5% of queries especially controversial ones, and are less consistent across repeated or slightly edited queries.
FlowBot automatically induces LLM workflows through bilevel optimization with textual gradients, achieving competitive performance against human-crafted baselines.
Discrete Tilt Matching recasts dLLM fine-tuning as state-level matching of tilted local unmasking posteriors, producing a stable weighted cross-entropy loss that improves Sudoku and Countdown performance when applied to LLaDA-8B-Instruct.
MuRGAt benchmark reveals that strong multimodal models frequently hallucinate citations in complex reasoning tasks despite correct answers, exposing a gap between internal reasoning and verifiable attribution.
Norm-Anchor Scaling breaks the norm-feedback loop in sequential LLM editing by anchoring value vectors to original norms, improving long-run performance by 72.2% and extending the editing horizon over 4x.
SSLogic uses LLM agents in a closed Generate-Validate-Refine loop to evolve 953 logic task families from 400 seeds, producing data that yields benchmark gains of +5.2 on SynLogic, +3.0 on AIME25, and +5.5 on BBH.
HiPRAG adds hierarchical process rewards to RL training for agentic RAG, reducing over-search to 2.3% and achieving 65.4-67.2% accuracy on seven QA benchmarks across 3B and 7B models.
SynBench benchmarks DP text generators across nine datasets and uses a new MIA to show that public pre-training on portions of private data overestimates synthetic text quality and breaks DP privacy bounds.
citing papers explorer
-
Trust but Verify: Prover-Verifier Deliberation for Selective LLM Prediction
Prover-verifier deliberation yields a high-confidence subset of LLM answers with ~30pp higher precision than the complement on GPQA Diamond by using defender-challenger dialogues.
-
Evaluating Cognitive Age Alignment in Interactive AI Agents
The paper presents ChildAgentEval as the first psychometrically grounded benchmark comparing MLLM-based agents' reasoning performance to age-specific human cognitive stages.
-
LGMT: Logic-Grounded Metamorphic Testing for Evaluating the Reasoning Reliability of LLMs
LGMT is a logic-grounded metamorphic testing framework that detects hidden reasoning defects in LLMs by checking consistency on semantically invariant inputs derived from FOL equivalences.
-
Scaling the Scaling Logic: Agentic Meta-Synthesis of Logic Reasoning
SSLogic uses LLM agents in a closed Generate-Validate-Refine loop to evolve 953 logic task families from 400 seeds, producing data that yields benchmark gains of +5.2 on SynLogic, +3.0 on AIME25, and +5.5 on BBH.
-
SynBench: A Benchmark for Differentially Private Text Generation
SynBench benchmarks DP text generators across nine datasets and uses a new MIA to show that public pre-training on portions of private data overestimates synthetic text quality and breaks DP privacy bounds.
-
Navigating Unreliable Parametric and Contextual Knowledge: Explicit Knowledge Conflict Resolution for LLM Inference
MACR adaptively assesses LLM confidence via semantic entropy then applies inductive multi-agent reasoning with rule-induction, conflict-analysis, and resolution agents to handle unreliable parametric and contextual knowledge.
-
TCP-MCP: Landscape-Guided Co-Evolution of Prompts and Communication Topologies for Multi-Agent Systems
TCP-MCP co-evolves prompts and topologies for multi-agent systems, reporting 82.66-96.61% accuracy on MMLU-Pro/MMLU/GSM8K while using up to 5.69x fewer tokens than debate baselines.
-
ECUAS$_n$: A family of metrics for principled evaluation of uncertainty-augmented systems
ECUAS_n is a parameterized family of proper scoring rules for jointly assessing prediction accuracy and uncertainty quality in automated decision systems.
-
The Shift Toward Open and Reproducible AI Research
Longitudinal study of 56,800 AI papers finds sixfold increase in code+data sharing from 2014-2024 with inferred reproducibility rising from 28% to 64%.
-
Neuro-Symbolic Verification of LLM Outputs for Data-Sensitive Domains (extended preprint)
Neuro-symbolic pipeline using formal logic and semantic embeddings detects hallucinations in LLM medical reports at 83%+ for entities and 72% for fabrications while cutting creation time 30%.
-
Context, Reasoning, and Hierarchy: A Cost-Performance Study of Compound LLM Agent Design in an Adversarial POMDP
In CybORG CAGE-2, programmatic state abstraction improves mean return up to 76% over raw observations while adding deliberation tools to hierarchies degrades performance up to 3.4x and increases token use.
-
When Helpfulness Becomes Sycophancy: Sycophancy is a Boundary Failure Between Social Alignment and Epistemic Integrity in Large Language Models
Sycophancy is a boundary failure between social alignment and epistemic integrity, captured by a three-condition framework plus taxonomy of targets, mechanisms, and severity.
-
pAI/MSc: ML Theory Research with Humans on the Loop
pAI/MSc is a customizable multi-agent system that reduces human steering by orders of magnitude when turning a hypothesis into a literature-grounded, mathematically established, experimentally supported manuscript draft in ML theory.
-
"Skill issues'': data-centric optimization of lakehouse agents
Data-centric optimization of skills for agents on a branching lakehouse improves accuracy by 31.9% on 25 tasks via state-verification evaluation.
- Explicit Logic Channel for Validation and Enhancement of MLLMs on Zero-Shot Tasks