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.
- 2026-07arXiv 2607.11020Can a Language Model Learn Facts Continually in Its Weights?
Charles O'Neill, Max Kirkby, Jonathon Liu et al. · Language models & NLP
Continual learning is a long-standing capability gap between LLMs and humans. Writing new knowledge into a model's weights routinely causes it to forget old knowledge, commonly denoted as "catastrophic forgetting".
- 2026-01arXiv 2601.03746Whose Facts Win? LLM Source Preferences under Knowledge Conflicts
Jakob Schuster, Vagrant Gautam, Katja Markert · Language models & NLP
As large language models (LLMs) are more frequently used in retrieval-augmented generation pipelines, it is increasingly relevant to study their behavior under knowledge conflicts.
- 2025-04arXiv 2504.12461On the Need to Rethink Trust in AI Assistants for Software Development: A Critical Review
Sebastian Baltes, Timo Speith, Brenda Chiteri et al. · Society, software & economics
Trust is a fundamental concept in human decision-making and collaboration that has long been studied in philosophy and psychology.
- 2024-06arXiv 2406.19354Fundamental Problems With Model Editing: How Should Rational Belief Revision Work in LLMs?
Peter Hase, Thomas Hofweber, Xiang Zhou et al. · Language models & NLP
The model editing problem concerns how language models should learn new facts about the world over time.
- 2024-05arXiv 2405.05904Does Fine-Tuning LLMs on New Knowledge Encourage Hallucinations?
Zorik Gekhman, Gal Yona, Roee Aharoni et al. · Language models & NLP
When large language models are aligned via supervised fine-tuning, they may encounter new factual information that was not acquired through pre-training.
Cited inStable wrong beliefs evade every self-check: a literature synthesis on model self-knowledge and confident errorsTo stop a model forgetting, rehearse what it knew. New context won't do it.Study: Consequence consistent updatesStudy: Continual learning order and companion dataStudy: Fortifying knowledge in continual learningStudy: Generational competitive continual learningStudy: Graded succession by earned authorityStudy: Logical coherence of LLM world modelsStudy: Model self knowledge and confident errorsStudy: Orchestrated specialists and bridge synthesisStudy: Stopping confident wrong expert votes - 2024-04arXiv 2404.13208The Instruction Hierarchy: Training LLMs to Prioritize Privileged Instructions
Eric Wallace, Kai Xiao, Reimar Leike et al. · Security
Today's LLMs are susceptible to prompt injections, jailbreaks, and other attacks that allow adversaries to overwrite a model's original instructions with their own malicious prompts.
- 2024-02arXiv 2402.11782What Evidence Do Language Models Find Convincing?
Alexander Wan, Eric Wallace, Dan Klein · Language models & NLP
Retrieval-augmented language models are being increasingly tasked with subjective, contentious, and conflicting queries such as "is aspartame linked to cancer".".
- 2023-12arXiv 2312.09390Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision
Collin Burns, Pavel Izmailov, Jan Hendrik Kirchner et al. · Language models & NLP
Widely used alignment techniques, such as reinforcement learning from human feedback (RLHF), rely on the ability of humans to supervise model behavior - for example, to evaluate whether a model faithfully followed instructions or generated safe outputs.
- 2023-11arXiv 2311.06305A Systematic Review on Fostering Appropriate Trust in Human-AI Interaction
Siddharth Mehrotra, Chadha Degachi, Oleksandra Vereschak et al. · Society, software & economics
Appropriate Trust in Artificial Intelligence (AI) systems has rapidly become an important area of focus for both researchers and practitioners. Various approaches have been used to achieve it, such as confidence scores, explanations, trustworthiness cues, or uncertainty communication.
- 2023-07arXiv 2307.12976Evaluating the Ripple Effects of Knowledge Editing in Language Models
Roi Cohen, Eden Biran, Ori Yoran et al. · Language models & NLP
Modern language models capture a large body of factual knowledge. However, some facts can be incorrectly induced or become obsolete over time, resulting in factually incorrect generations.
Cited inA gradient kernel flags contradictions but does not predict forgetting, and steering rehearsal by it backfiresCoherence is selected, anchored, then pruned: a literature synthesis on logical coherence in LLM world modelsEWC protects what its Fisher sees: a literature synthesis on elastic weight consolidation, scale and coherenceIterated self-closure settles fast, but only evidence that overrides self-belief makes the settled state trueTo stop a model forgetting, rehearse what it knew. New context won't do it.Study: Consequence consistent updatesStudy: Continual learning order and companion dataStudy: Decision signatures for conflict detectionStudy: Elastic weight consolidation deep diveStudy: Graded succession by earned authorityStudy: Hardware guardian core for continual learningStudy: Logical coherence of LLM world modelsStudy: Orchestrated specialists and bridge synthesis - 2023-05arXiv 2305.14795MQuAKE: Assessing Knowledge Editing in Language Models via Multi-Hop Questions
Zexuan Zhong, Zhengxuan Wu, Christopher D. Manning et al. · Language models & NLP
The information stored in large language models (LLMs) falls out of date quickly, and retraining from scratch is often not an option. This has recently given rise to a range of techniques for injecting new facts through updating model weights.
Cited inTrust the premises, not the confidence: which belief to drop when an LLM contradicts itself, and why the solver is not the bottleneckStudy: Consequence consistent updatesStudy: Continual learning order and companion dataStudy: Graded succession by earned authorityStudy: LLMs modifying open model weightsStudy: Logical coherence of LLM world modelsStudy: Orchestrated specialists and bridge synthesisStudy: Weight editing to prevent forgetting - 2023-05arXiv 2305.13300Adaptive Chameleon or Stubborn Sloth: Revealing the Behavior of Large Language Models in Knowledge Conflicts
Jian Xie, Kai Zhang, Jiangjie Chen et al. · Language models & NLP
By providing external information to large language models (LLMs), tool augmentation (including retrieval augmentation) has emerged as a promising solution for addressing the limitations of LLMs' static parametric memory.
Cited inCoherence is selected, anchored, then pruned: a literature synthesis on logical coherence in LLM world modelsSuccessors should earn authority, not win contests: graded succession in continual learningStudy: Decision signatures for conflict detectionStudy: Graded succession by earned authorityStudy: Logical coherence of LLM world modelsStudy: Memory world models and source trustStudy: Model self knowledge and confident errors - 2022-10arXiv 2210.01948Game-theoretic statistics and safe anytime-valid inference
Aaditya Ramdas, Peter Grünwald, Vladimir Vovk et al. · Statistics & theory
Safe anytime-valid inference (SAVI) provides measures of statistical evidence and certainty -- e-processes for testing and confidence sequences for estimation -- that remain valid at all stopping times, accommodating continuous monitoring and analysis of accumulating data and optional stopping or continuation for…
- 2022-03arXiv 2203.05482Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time
Mitchell Wortsman, Gabriel Ilharco, Samir Yitzhak Gadre et al. · Machine learning
The conventional recipe for maximizing model accuracy is to (1) train multiple models with various hyperparameters and (2) pick the individual model which performs best on a held-out validation set, discarding the remainder.
- 2022-03arXiv 2203.02155Training language models to follow instructions with human feedback
Long Ouyang, Jeff Wu, Xu Jiang et al. · Language models & NLP
Making language models bigger does not inherently make them better at following a user's intent. For example, large language models can generate outputs that are untruthful, toxic, or simply not helpful to the user.
- 2021-09arXiv 2109.05052Entity-Based Knowledge Conflicts in Question Answering
Shayne Longpre, Kartik Perisetla, Anthony Chen et al. · Language models & NLP
Knowledge-dependent tasks typically use two sources of knowledge: parametric, learned at training time, and contextual, given as a passage at inference time.
- 2021-09arXiv 2109.01903Robust fine-tuning of zero-shot models
Mitchell Wortsman, Gabriel Ilharco, Jong Wook Kim et al. · Vision
, without fine-tuning on a specific dataset). Although existing fine-tuning methods substantially improve accuracy on a given target distribution, they often reduce robustness to distribution shifts.
- 2021-06arXiv 2106.15110Time-Aware Language Models as Temporal Knowledge Bases
Bhuwan Dhingra, Jeremy R. Cole, Julian Martin Eisenschlos et al. · Transactions of the Association for Computational Linguistics 2022; 10 257-273 · Language models & NLP
Many facts come with an expiration date, from the name of the President to the basketball team Lebron James plays for.
- 2019-09arXiv 1909.02187Adaptive and Efficient Algorithms for Tracking the Best Expert
Shiyin Lu, Lijun Zhang · Machine learning
In this paper, we consider the problem of prediction with expert advice in dynamic environments. We choose tracking regret as the performance metric and develop two adaptive and efficient algorithms with data-dependent tracking regret bounds.
- 2018-10arXiv 1810.08240Time-uniform, nonparametric, nonasymptotic confidence sequences
Steven R. Howard, Aaditya Ramdas, Jon McAuliffe et al. · Ann. Statist. 49(2): 1055-1080 (April 2021) · Statistics & theory
A confidence sequence is a sequence of confidence intervals that is uniformly valid over an unbounded time horizon. Our work develops confidence sequences whose widths go to zero, with nonasymptotic coverage guarantees under nonparametric conditions.
- 2017-07arXiv 1707.06347Proximal Policy Optimization Algorithms
John Schulman, Filip Wolski, Prafulla Dhariwal et al. · Machine learning
We propose a new family of policy gradient methods for reinforcement learning, which alternate between sampling data through interaction with the environment, and optimizing a "surrogate" objective function using stochastic gradient ascent.
- 2017-04arXiv 1704.02030Using stacking to average Bayesian predictive distributions
Yuling Yao, Aki Vehtari, Daniel Simpson et al. · Statistics & theory
The widely recommended procedure of Bayesian model averaging is flawed in the M-open setting in which the true data-generating process is not one of the candidate models being fit.
- 2017-03arXiv 1703.01780Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results
Antti Tarvainen, Harri Valpola · Neuroscience & neural computation
The recently proposed Temporal Ensembling has achieved state-of-the-art results in several semi-supervised learning benchmarks. It maintains an exponential moving average of label predictions on each training example, and penalizes predictions that are inconsistent with this target.
- 2016-06arXiv 1606.09282Learning without Forgetting
Zhizhong Li, Derek Hoiem · Vision
When building a unified vision system or gradually adding new capabilities to a system, the usual assumption is that training data for all tasks is always available.
- 2016-06arXiv 1606.04671Progressive Neural Networks
Andrei A. Rusu, Neil C. Rabinowitz, Guillaume Desjardins et al. · Machine learning
Learning to solve complex sequences of tasks--while both leveraging transfer and avoiding catastrophic forgetting--remains a key obstacle to achieving human-level intelligence.
- 2015-12arXiv 1512.04922Always Valid Inference: Bringing Sequential Analysis to A/B Testing
Ramesh Johari, Leo Pekelis, David J. Walsh · Statistics & theory
A/B tests are typically analyzed via frequentist p-values and confidence intervals; but these inferences are wholly unreliable if users endogenously choose samples sizes by *continuously monitoring* their tests.
- 2015-05arXiv 1505.02463A Survey on Truth Discovery
Yaliang Li, Jing Gao, Chuishi Meng et al. · Retrieval & data
Thanks to information explosion, data for the objects of interest can be collected from increasingly more sources. However, for the same object, there usually exist conflicts among the collected multi-source information.
- 2015-02arXiv 1502.05477Trust Region Policy Optimization
John Schulman, Sergey Levine, Philipp Moritz et al. · Machine learning
We describe an iterative procedure for optimizing policies, with guaranteed monotonic improvement. By making several approximations to the theoretically-justified procedure, we develop a practical algorithm, called Trust Region Policy Optimization (TRPO).
- 2015-02arXiv 1502.03519Knowledge-Based Trust: Estimating the Trustworthiness of Web Sources
Xin Luna Dong, Evgeniy Gabrilovich, Kevin Murphy et al. · Retrieval & data
The quality of web sources has been traditionally evaluated using exogenous signals such as the hyperlink structure of the graph. We propose a new approach that relies on endogenous signals, namely, the correctness of factual information provided by the source.
- 2010-08arXiv 1008.4532Switching between Hidden Markov Models using Fixed Share
Wouter M. Koolen, Tim van Erven · Machine learning
In prediction with expert advice the goal is to design online prediction algorithms that achieve small regret (additional loss on the whole data) compared to a reference scheme.
- 2008-07arXiv 0807.1005Catching Up Faster by Switching Sooner: A Prequential Solution to the AIC-BIC Dilemma
Tim van Erven, Peter Grunwald, Steven de Rooij · Statistics & theory
Bayesian model averaging, model selection and its approximations such as BIC are generally statistically consistent, but sometimes achieve slower rates og convergence than other methods such as AIC and leave-one-out cross-validation.
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.
