NeuroGenesysThe research we build on.
1,514 papers by other researchers that our work cites, from continual learning and verification to language models, agents and hardware. Each entry links to the paper on arXiv and to where we use it. 301 are cited on published pages of this site; the rest in the research notes behind our studies. Our own papers are in the archive.
- 2026-09arXiv 2609.07395Uncertainty Quantification for LLM Agents: A Taxonomy, an Evaluation Protocol, and an Empirical Study
Moule Lin, Qizhen Lan, Shuhao Guan et al. · Retrieval & data
Large language models (LLMs) are no longer deployed only for single-turn conversation but increasingly act as agents that plan, call tools, retrieve evidence, maintain memory, and interact over long horizons, often together with other agents through multi-turn conversations.
- 2026-09arXiv 2609.03900Beyond Endpoint Scores: Time- and Capacity-Conditioned Evaluation of Continual Knowledge Updating
Heejin Choi · Machine learning
Continual knowledge-updating methods are often declared superior from one final checkpoint and one conventional adapter rank. We show that this can be insufficient to identify the better operating point.
- 2026-09arXiv 2609.04445Conformity Breaks Conformal Prediction
Yibo Hu, Hanyu Su · Machine learning
A conformal certificate can be valid when an LLM answers alone and invalid when the same LLM sees peers that unanimously assert a wrong answer. The question is unchanged; the model's score for the correct answer changes.
- 2026-09arXiv 2609.02548Learn from Whoever Is Right: Answer-Verified Multi-Teacher Distillation for Multi-Domain LLMs
Xixiang He, Xingming Li, Baiqi Wu et al. · Machine learning
Modern large language models (LLMs) rely on reinforcement learning to build strong capabilities in individual domains, but integrating those capabilities into a single deployable model remains challenging.
- 2026-09arXiv 2609.02849Post-Training Language Models for Gold-Medal Performance in Coding Competitions
Aleksander Ficek, Sean Narenthiran, Mehrzad Samadi et al. · Machine learning
Competitive programming has become a key test of large language model reasoning, with international competitions such as IOI and ICPC representing its most challenging settings.
Cited inStudy: Superintelligence - 2026-09arXiv 2609.22218Toollery: Scaling LLM Agents to Thousands of Skills and Tools
Xiangxi Tian, Ran Guan · Machine learning
As LLM agents are exposed to hundreds to tens of thousands of skills, tools, and API functions, full-library prompting becomes costly, slow, and less reliable: each added candidate increases prompt tokens and latency, while longer candidate lists introduce more distractors for LLM selection.
Cited inInternal notes - 2026-09arXiv 2609.01852The Memory Trust Gap: Capability-Dependent Failures in Persistent-Memory Agents
Jundong Hu, Shekar Ramachandran · AI & agents
Persistent memory supports personalized agents, but a stale stored fact can override current authoritative evidence without warning. We study when this harm begins as model capability changes.
Cited inInternal notes - 2026-08arXiv 2608.30661SwarmBench: Can Large Language Models Act as Agent Swarm Orchestrators?
Jinshan Gao, Zhuoran Jin, Tianyi Men et al. · Language models & NLP
Large language model-based multi-agent systems are evolving from fixed interaction topologies toward dynamically orchestrated Agent Swarms.
- 2026-08arXiv 2609.00188ZimaBlue: Evolving Generalizable World Action Models through Scalable Video Pre-training
Xionghao Wu, Yijun Yang, Shiyang Zhou et al. · Vision
Robotic manipulation faces a fundamental scaling challenge: robust generalization demands broad physical experience, yet action-labeled robot trajectories are expensive to collect and inherently limited in diversity.
- 2026-08arXiv 2608.26623AgentJudgeBench: A Multi-Difficulty Benchmark for Evaluating LLM Judges on Agentic Tool-Calling
Abhigya Verma, Amit Kumar Saha, Seganrasan Subramanian et al. · AI & agents
LLM judges are widely used to evaluate agentic tool-calling systems, yet their reliability on structured, dependency-driven workflows remains largely unexamined.
Cited inInternal notes - 2026-08arXiv 2608.27046Performance Foundations of Parallel & Distributed Reasoning Language Models
Maciej Besta, Leonard Schmidt, Lara Nonino et al. · Machine learning
Reinforcement Learning with Verifiable Rewards (RLVR) and other RL-style post-training paradigms have been used for aligning large language models (LLMs) with reasoning standards.
- 2026-08arXiv 2608.27427Persona-Execution Separation: An Architecture Pattern for Evolving LLM Agents under Execution Audit
Yisen Xi · Society, software & economics
Large language model (LLM) agents in governed organizations must let the persona (instructions, tone, self-presentation) evolve freely, while keeping execution (stateful, audited work) traceable.
- 2026-08arXiv 2608.25429Distance Is Not Enough: Forget-Retain Alignment Gap Predicts LLM Relearning Robustness
Yi Chen, Hanna Hsieh, Shuhong Liu et al. · Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP 2026), pages 44542-44555 · AI & agents
Machine unlearning aims to make a model forget specific data, yet unlearned LLMs often fail to stay unlearned: brief fine-tuning can revive removed knowledge.
- 2026-08arXiv 2608.26292On Scope Classification and Current Knowledge-Editing Benchmarks: A Negative Result, with INLAY as a Gradient-Free Case Study
Aditya Pratap Singh · Language models & NLP
Every memory-based knowledge editor in the SERAC lineage depends on a scope decision: given a query, does a stored edit apply? We report that current knowledge-editing benchmarks cannot measure this decision at all.
- 2026-08arXiv 2609.22170Multiple latent orderings better predict language model preferences
Aviral Chawla, William H. W. Thompson, Jean-Gabriel Young · Machine learning
Language models are frequently employed in settings where they are asked to make value judgments and choices. These observed choices often exhibit intransitivity: A model may prefer item to and to , while also preferring to .
- 2026-08arXiv 2608.22884Predicting the scale limits of social mechanisms in agent societies
Zengqing Wu, Chuan Xiao · AI & agents
Societies of interacting language-model agents offer a controllable and repeatable way to study collective behaviour at scales that would be difficult to test with people.
- 2026-08arXiv 2608.22215Dual-Layer Agentic Memory with Fast Write Routing and Slow Consolidation
Wenzhi Li, Dong Nie, Rui Lan et al. · Language models & NLP
Large language model (LLM) agents operate in dynamic environments where knowledge continuously evolves. Existing memory systems typically treat external memory as a monotonically growing repository, inevitably leading to retrieval degradation and increasing computational costs over time.
- 2026-08arXiv 2609.27954When Accuracy Gaps Fail to Certify: Auditing Cross-Domain Recalibration of LLM Judges
Fariya Afrin, Ibne Farabi Shihab · Machine learning
A scalar recalibration map fitted for an LLM judge on one task can fail when the task distribution changes, but the source-target accuracy gap is often treated as a proxy for that failure.
- 2026-08arXiv 2608.16141Continual-learning rules shape representational drift
Yikai Si, Shanshan Qin · Neuroscience & neural computation
Lifelong learning requires acquiring new knowledge without erasing the old. Yet neural population codes for familiar stimuli and behaviors change over days and weeks.
- 2026-08arXiv 2608.15901Layers Matter: Why Continual Learning Regularization Should Be Layer-Adaptive
Brian B. Moser, Ahmed Anwar, Tobias Christian Nauen et al. · Machine learning
Continual learning regularizers like EWC fight forgetting by penalizing changes from previous-task parameters with per-parameter importance, typically diagonal Fisher values.
- 2026-08arXiv 2608.13921When Personal Memory Has No Single Answer: Evaluating LLM Agents under Irreducible Conflict
Lu Yang, Shusheng Xu, Zhuoran Li et al. · AI & agents
LLM agents increasingly maintain personal memory across sessions, but it can conflict. Preferences depend on context, behavior evolves, and sources can conflict.
Cited inInternal notes - 2026-08arXiv 2608.13607No Universal Signal Predicts Sample-Level LLM Regression under Version Updates
Jia Sheng, Yiwei Lu · AI & agents
Frontier LLMs are updated frequently and typically outperform their predecessors in aggregate. But aggregate gains say little about individual samples: an update can still cause sample-level regression, where a response correct under the old model becomes incorrect under the new one.
- 2026-08arXiv 2608.09819Macaron-V1: Towards Open Continual Learning with Self-Improvement and Mixture-of-LoRA
Mind Lab, :, Vin Bo et al. · Machine learning
Macaron-V1 is an open agent-model family for experiential intelligence: learning from experience in real environments and continuing to learn after deployment. It is organized around two system goals.
- 2026-08arXiv 2608.08650The Evolution of Mixture-of-Experts Architectures in Large Language Models: Routing, Topology, Load Balancing, and Expert Parallelism
Jiguo Li · Language models & NLP
Mixture-of-Experts models increase parameter capacity while keeping the computation activated by each token bounded, but their architectural evolution cannot be explained by a chronological list of model releases alone.
- 2026-08arXiv 2608.08002Evaluator Ensembles Under Reward Hacking: Covariance Geometry and Finite-Search Guarantees
Fariya Afrin, Ibne Farabi Shihab · Machine learning
Language-model judges and reward models enable scalable supervision, but finite optimization can exploit evaluator errors rather than improve response quality. We characterize this failure through the covariance geometry of evaluator ensembles.
- 2026-08arXiv 2608.08239The Replay Gap: Static Evaluation of Model Switching in LLM Agents Scores the Wrong World
Ashritha Gonuguntla · Machine learning
LLM routers promise efficiency by matching each request to the cheapest adequate model, and are increasingly applied per step inside multi-step agents.
- 2026-08arXiv 2608.07429TEPA: Revoking Stale Memories for Conflict-Robust Language Agents
Yan Zhou, Yue Ouyang, Kaiyang Zheng et al. · AI & agents
Long-term memory enables language agents to reuse past facts, preferences, and task experience. Persistence also creates a central falsifiability problem: when the world changes, stale memories can remain retrievable and pollute the prompt.
Cited inInternal notes - 2026-08arXiv 2608.07622Controlled Memory Interference in Continual LLM Agents
Ao Ding, Hongzong LI, Shiqin Tang et al. · AI & agents
Long-term memory enables AI agents to maintain continuity across sessions, personalize behavior, and evolve through accumulated experience. Yet memory evolution is not simply a process of storing more information: new experiences may reinforce, revise, or interfere with existing memory states.
Cited inInternal notes - 2026-08arXiv 2608.05499APQF: Agentic Profiling-Guided Structured Pruning and Mixed-Precision Quantization with Adaptive Fine-Tuning
Sadegh Jafari, Mohiuddin Bilwal, Fan Zhou et al. · Vision
Modern deep neural networks achieve strong performance, but their scale makes them costly and slow, especially on resource-constrained edge devices. Pruning and quantization address this, but rely on manual, expert choices and on algorithms that are hard to apply across architectures.
- 2026-08arXiv 2608.06216Continual Learning in Transition
Zhiyan Hou, Dan Zhang, Tao Feng et al. · Machine learning
, training strategies, architectural designs, and weight adaptation. However, emerging paradigms are reshaping the scope of CL beyond this traditional model adaptation view.
Cited inInternal notes - 2026-08arXiv 2608.04454Beyond Global Routing Aggregation: Phase-Aware Expert Merging for MoE Vision-Language Models
Hongyu Zhang, Cheng Yan, Xiang Xia et al. · Vision
Mixture-of-experts vision-language models (MoE-VLMs) increase model capacity with sparse expert activation, yet deployment requires storing the full expert pool.
- 2026-08arXiv 2609.20846Rewarding Efficient Reasoning Improves Abstention on Underspecified Tasks in Reasoning Models
Polina Tsvilodub, Max Höth, Michael Franke et al. · Language models & NLP
While modern large reasoning models (LRMs) excel at providing correct answers in many tasks, we provide additional evidence that they often struggle with a critical capability: knowing when to abstain from answering.
- 2026-08arXiv 2608.01619When Memory Updates but Behavior Does Not: Repairing Implicit Stale Dependencies in Personalized Agent Responses
Haofei Sun, Lin He · AI & agents
Memory-augmented agents can know that a user's stored state is outdated and still plan around the old value. The STALE benchmark calls this the implicit policy adaptation (IPA) gap.
Cited inInternal notes - 2026-08arXiv 2608.02528Uncertainty Is Not Enough: Value-of-Information Routing for Mixtures of LoRA Experts
Tom Saliencro, Rohan Desai, Priya Nair et al. · Machine learning
Mixtures of low-rank adaptation experts increase parameter-efficient capacity by routing each input through a subset of adapters. Recent dynamic routers activate more experts when the router or prediction is uncertain.
- 2026-08arXiv 2608.02758Everyone Conforms, No One Believes: Pluralistic Ignorance in LLM Agent Populations
Yashwanth YS · AI & agents
LLM-based multi-agent systems are increasingly used to simulate social dynamics, from opinion formation to collective decision-making.
- 2026-08arXiv 2608.02827Emergence of Biased Consensus in Multi-Agent LLM Debates
Maya Okawa · AI & agents
Multi-agent LLM debates achieve strong performance on decision-making tasks as well as problem-solving benchmarks, yet their safety and fairness risks remain poorly understood.
- 2026-08arXiv 2608.01437Beyond Routing Saturation: A Long-Horizon Class-Incremental Perspective on Expert Routing in Multimodal Continual Instruction Tuning
Huiyu Yi, Yongqi Xu, Bogang Zhang et al. · AI & agents
Multimodal Continual Instruction Tuning (MCIT) enables multimodal large language models to acquire new tasks sequentially while retaining previously learned capabilities.
- 2026-07arXiv 2608.00301Abstention as an Action Can Kill Both the Reward Gradient and the KL Anchor: Collapse Law and Repair for Error-Penalized Reinforcement Learning
Xujun Che, Yuchen Yuan, Weida Zhao et al. · Machine learning
Error-penalized scoring rules ( for a correct answer, for a wrong one, $0$ for abstaining) are increasingly prescribed against hallucination: a rational agent facing such a rule answers exactly when its correctness probability exceeds Chow's threshold .
- 2026-07arXiv 2607.28308Beyond Geometric Complementarity: Coherent Overlap in Sparse Mixture-of-Experts Routing
Huiyuan Tian, Bonan Xu, Shijian Li · Machine learning
Sparse mixture-of-experts (MoE) language models route each token to multiple experts, suggesting a geometric account of their benefit: co-selected experts should contribute distinct representation directions.
- 2026-07arXiv 2607.27036Mitigating Compounding Error via Video Representation Regularization
Taiye Chen, Qi Zhang, Yisen Wang · Vision
Video diffusion-based world models enable long autoregressive video generation for robotics, autonomous driving and simulation tasks, yet sliding-window autoregressive inference suffers from severe error accumulation that degrades frame quality over time.
Built from every arXiv paper cited in our papers, posts, topic hubs, reports and research notes, with titles, authors, dates and abstracts from the arXiv API (updated 2026-11-12). Summaries are the opening sentences of each paper's own abstract, not our assessment; a citation means we used or checked the work, not that we endorse it. Fields follow each paper's primary arXiv category. Papers cited only outside arXiv (journals, proceedings) are not listed yet. Entries are generated from our citations, so a mistyped ID in a note would list the wrong paper; we remove those as we find them. Report a wrong entry.
