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.
- 2024-12arXiv 2412.14711ReMoE: Fully Differentiable Mixture-of-Experts with ReLU Routing
Ziteng Wang, Jun Zhu, Jianfei Chen · Machine learning
Sparsely activated Mixture-of-Experts (MoE) models are widely adopted to scale up model capacity without increasing the computation budget. However, vanilla TopK routers are trained in a discontinuous, non-differentiable way, limiting their performance and scalability.
- 2024-12arXiv 2412.14093Alignment faking in large language models
Ryan Greenblatt, Carson Denison, Benjamin Wright et al. · AI & agents
We present a demonstration of a large language model engaging in alignment faking: selectively complying with its training objective in training to prevent modification of its behavior out of training.
- 2024-12arXiv 2412.12472Knowledge Boundary of Large Language Models: A Survey
Moxin Li, Yong Zhao, Wenxuan Zhang et al. · Language models & NLP
Although large language models (LLMs) store vast amount of knowledge in their parameters, they still have limitations in the memorization and utilization of certain knowledge, leading to undesired behaviors such as generating untruthful and inaccurate responses.
- 2024-12arXiv 2412.09565Obfuscated Activations Bypass LLM Latent-Space Defenses
Luke Bailey, Alex Serrano, Abhay Sheshadri et al. · Machine learning
Recent latent-space monitoring techniques have shown promise as defenses against LLM attacks. These defenses act as scanners that seek to detect harmful activations before they lead to undesirable actions.
- 2024-12arXiv 2412.09764Memory Layers at Scale
Vincent-Pierre Berges, Barlas Oğuz, Daniel Haziza et al. · Language models & NLP
Memory layers use a trainable key-value lookup mechanism to add extra parameters to a model without increasing FLOPs. Conceptually, sparsely activated memory layers complement compute-heavy dense feed-forward layers, providing dedicated capacity to store and retrieve information cheaply.
- 2024-12arXiv 2412.06117Paraconsistent Belief Revision: A Replacement-Enriched LFI for Epistemic Entrenchment
Marcelo E. Coniglio, Martin Figallo, Rafael R. Testa · Other
We further develop the formal foundations of Paraconsistent Belief Revision (PBR) by introducing Logics of Formal Inconsistency (LFIs) specifically designed to support the development of epistemic entrenchment-based models for belief change.
- 2024-12arXiv 2412.06538Understanding Factual Recall in Transformers via Associative Memories
Eshaan Nichani, Jason D. Lee, Alberto Bietti · Machine learning
Large language models have demonstrated an impressive ability to perform factual recall. Prior work has found that transformers trained on factual recall tasks can store information at a rate proportional to their parameter count.
- 2024-12arXiv 2412.06559ProcessBench: Identifying Process Errors in Mathematical Reasoning
Chujie Zheng, Zhenru Zhang, Beichen Zhang et al. · AI & agents
As language models regularly make mistakes when solving math problems, automated identification of errors in the reasoning process becomes increasingly significant for their scalable oversight.
- 2024-12arXiv 2412.05346BadGPT-4o: stripping safety finetuning from GPT models
Ekaterina Krupkina, Dmitrii Volkov · Security
2023's simple fine-tuning poisoning technique strips GPT-4o's safety guardrails without degrading the model. The BadGPT attack matches best white-box jailbreaks on HarmBench and StrongREJECT.
- 2024-12arXiv 2412.04139Monet: Mixture of Monosemantic Experts for Transformers
Jungwoo Park, Young Jin Ahn, Kee-Eung Kim et al. · AI & agents
Understanding the internal computations of large language models (LLMs) is crucial for aligning them with human values and preventing undesirable behaviors like toxic content generation.
- 2024-12arXiv 2412.03671Tight Lower Bounds and Improved Convergence in Performative Prediction
Pedram Khorsandi, Rushil Gupta, Mehrnaz Mofakhami et al. · Machine learning
Performative prediction is a framework accounting for the shift in the data distribution induced by the prediction of a model deployed in the real world.
- 2024-12arXiv 2412.01004Take Only What You Need: Rank Minimization as an Implicit Forgetting Regularizer in Continual Learning
Haodong Lu, Chongyang Zhao, Minhui Xue et al. · Vision
The central tension in continual learning (CL) is the trade-off between plasticity (acquiring new knowledge) and stability (retaining prior knowledge).
- 2024-11arXiv 2411.17367Efficient transformer adaptation for analog in-memory computing via low-rank adapters
Chen Li, Elena Ferro, Corey Lammie et al. · Neuromorphic Computing and Engineering 6, 014011 (2026) · Hardware, photonics & physics
Analog In-Memory Computing (AIMC) offers a promising solution to the von Neumann bottleneck. However, deploying transformer models on AIMC remains challenging due to their inherent need for flexibility and adaptability across diverse tasks.
- 2024-11arXiv 2411.17501The Limits of Inference Scaling Through Resampling
Benedikt Stroebl, Sayash Kapoor, Arvind Narayanan · Machine learning
Recent research has generated hope that inference scaling, such as resampling solutions until they pass verifiers like unit tests, could allow weaker models to match stronger ones.
- 2024-11arXiv 2411.17766Dual Prototypes for Adaptive Pre-Trained Model in Class-Incremental Learning
Zhiming Xu, Suorong Yang, Baile Xu et al. · Machine learning
Class-incremental learning (CIL) aims to learn new classes while retaining previous knowledge. Although pre-trained model (PTM) based approaches show strong performance, directly fine-tuning PTMs on incremental task streams often causes renewed catastrophic forgetting.
- 2024-11arXiv 2411.15399Less is More: Optimizing Function Calling for LLM Execution on Edge Devices
Varatheepan Paramanayakam, Andreas Karatzas, Iraklis Anagnostopoulos et al. · Other
The advanced function-calling capabilities of foundation models open up new possibilities for deploying agents to perform complex API tasks. However, managing large amounts of data and interacting with numerous APIs makes function calling hardware-intensive and costly, especially on edge devices.
Cited inStudy: Genesys program prior art - 2024-11arXiv 2411.15114RE-Bench: Evaluating frontier AI R&D capabilities of language model agents against human experts
Hjalmar Wijk, Tao Lin, Joel Becker et al. · Machine learning
Frontier AI safety policies highlight automation of AI research and development (R&D) by AI agents as an important capability to anticipate. However, there exist few evaluations for AI R&D capabilities, and none that are highly realistic and have a direct comparison to human performance.
- 2024-11arXiv 2411.14257Do I Know This Entity? Knowledge Awareness and Hallucinations in Language Models
Javier Ferrando, Oscar Obeso, Senthooran Rajamanoharan et al. · Language models & NLP
Hallucinations in large language models are a widespread problem, yet the mechanisms behind whether models will hallucinate are poorly understood, limiting our ability to solve this problem.
- 2024-11arXiv 2411.13619Non-Linear Outlier Synthesis for Out-of-Distribution Detection
Lars Doorenbos, Raphael Sznitman, Pablo Márquez-Neila · Vision
The reliability of supervised classifiers is severely hampered by their limitations in dealing with unexpected inputs, leading to great interest in out-of-distribution (OOD) detection.
- 2024-11arXiv 2411.10109LLM Agents Grounded in Self-Reports Enable General-Purpose Simulation of Individuals
Joon Sung Park, Carolyn Q. Zou, Jonne Kamphorst et al. · AI & agents
Machine learning can predict human behavior well when substantial structured data are available for well-defined outcomes. Such models are typically outcome-specific, however, requiring training data for each target outcome, limiting their applicability to new domains.
- 2024-11arXiv 2411.07175Continual Memorization of Factoids in Language Models
Howard Chen, Jiayi Geng, Adithya Bhaskar et al. · Transactions on Machine Learning Research, 2026 · Language models & NLP
As new knowledge rapidly accumulates, language models (LMs) with pretrained knowledge quickly become obsolete. A common approach to updating LMs is fine-tuning them directly on new knowledge.
- 2024-11arXiv 2411.04368Measuring short-form factuality in large language models
Jason Wei, Nguyen Karina, Hyung Won Chung et al. · Language models & NLP
We present SimpleQA, a benchmark that evaluates the ability of language models to answer short, fact-seeking questions. We prioritized two properties in designing this eval.
- 2024-11arXiv 2411.04468Magentic-One: A Generalist Multi-Agent System for Solving Complex Tasks
Adam Fourney, Gagan Bansal, Hussein Mozannar et al. · AI & agents
Modern AI agents, driven by advances in large foundation models, promise to enhance our productivity and transform our lives by augmenting our knowledge and capabilities.
- 2024-11arXiv 2411.02385How Far is Video Generation from World Model: A Physical Law Perspective
Bingyi Kang, Yang Yue, Rui Lu et al. · Vision
OpenAI's Sora highlights the potential of video generation for developing world models that adhere to fundamental physical laws. However, the ability of video generation models to discover such laws purely from visual data without human priors can be questioned.
- 2024-11arXiv 2411.02454Graph-based Confidence Calibration for Large Language Models
Yukun Li, Sijia Wang, Lifu Huang et al. · Language models & NLP
Reliable confidence estimation is essential for enhancing the trustworthiness of large language models (LLMs), especially in high-stakes scenarios. Despite its importance, accurately estimating confidence in LLM responses remains a significant challenge.
- 2024-11arXiv 2411.01111Rule Based Rewards for Language Model Safety
Tong Mu, Alec Helyar, Johannes Heidecke et al. · AI & agents
Reinforcement learning based fine-tuning of large language models (LLMs) on human preferences has been shown to enhance both their capabilities and safety behavior.
- 2024-11arXiv 2411.00430Class Incremental Learning with Task-Specific Batch Normalization and Out-of-Distribution Detection
Zhiping Zhou, Xuchen Xie, Yiqiao Qiu et al. · Machine learning
This study focuses on incremental learning for image classification, exploring how to reduce catastrophic forgetting of all learned knowledge when access to old data is restricted.
- 2024-10arXiv 2411.00114Project Sid: Many-agent simulations toward AI civilization
Altera. AL, Andrew Ahn, Nic Becker et al. · AI & agents
AI agents have been evaluated in isolation or within small groups, where interactions remain limited in scope and complexity. Large-scale simulations involving many autonomous agents -- reflecting the full spectrum of civilizational processes -- have yet to be explored.
- 2024-10arXiv 2410.21819Self-Preference Bias in LLM-as-a-Judge
Koki Wataoka, Tsubasa Takahashi, Ryokan Ri · Language models & NLP
Automated evaluation leveraging large language models (LLMs), commonly referred to as LLM evaluators or LLM-as-a-judge, has been widely used in measuring the performance of dialogue systems.
- 2024-10arXiv 2410.20783Graph-based Uncertainty Metrics for Long-form Language Model Outputs
Mingjian Jiang, Yangjun Ruan, Prasanna Sattigeri et al. · Language models & NLP
Recent advancements in Large Language Models (LLMs) have significantly improved text generation capabilities, but these systems are still known to hallucinate, and granular uncertainty estimation for long-form LLM generations remains challenging.
- 2024-10arXiv 2410.21228LoRA vs Full Fine-tuning: An Illusion of Equivalence
Reece Shuttleworth, Jacob Andreas, Antonio Torralba et al. · Machine learning
Fine-tuning is a crucial paradigm for adapting pre-trained large language models to downstream tasks. Recently, methods like Low-Rank Adaptation (LoRA) have been shown to effectively fine-tune LLMs with an extreme reduction in trainable parameters.
- 2024-10arXiv 2410.20526Llama Scope: Extracting Millions of Features from Llama-3.1-8B with Sparse Autoencoders
Zhengfu He, Wentao Shu, Xuyang Ge et al. · Machine learning
Sparse Autoencoders (SAEs) have emerged as a powerful unsupervised method for extracting sparse representations from language models, yet scalable training remains a significant challenge.
- 2024-10arXiv 2410.20098Self-Normalized Resets for Plasticity in Continual Learning
Vivek F. Farias, Adam D. Jozefiak · Machine learning
Plasticity Loss is an increasingly important phenomenon that refers to the empirical observation that as a neural network is continually trained on a sequence of changing tasks, its ability to adapt to a new task diminishes over time.
- 2024-10arXiv 2410.19278Applying sparse autoencoders to unlearn knowledge in language models
Eoin Farrell, Yeu-Tong Lau, Arthur Conmy · Machine learning
We investigate whether sparse autoencoders (SAEs) can be used to remove knowledge from language models. We use the biology subset of the Weapons of Mass Destruction Proxy dataset and test on the gemma-2b-it and gemma-2-2b-it language models.
- 2024-10arXiv 2410.18785Should We Really Edit Language Models? On the Evaluation of Edited Language Models
Qi Li, Xiang Liu, Zhenheng Tang et al. · AI & agents
Model editing has become an increasingly popular alternative for efficiently updating knowledge within language models. Current methods mainly focus on reliability, generalization, and locality, with many methods excelling across these criteria.
- 2024-10arXiv 2410.16713Collapse or Thrive? Perils and Promises of Synthetic Data in a Self-Generating World
Joshua Kazdan, Rylan Schaeffer, Apratim Dey et al. · Machine learning
What happens when generative machine learning models are pretrained on web-scale datasets containing data generated by earlier models? collapse can be avoided) by managing how available data are used in pretraining.
- 2024-10arXiv 2410.17146LiNeS: Post-training Layer Scaling Prevents Forgetting and Enhances Model Merging
Ke Wang, Nikolaos Dimitriadis, Alessandro Favero et al. · Machine learning
Fine-tuning pre-trained models has become the standard approach to endow them with specialized knowledge, but it poses fundamental challenges.
- 2024-10arXiv 2410.17413Scalable Influence and Fact Tracing for Large Language Model Pretraining
Tyler A. Chang, Dheeraj Rajagopal, Tolga Bolukbasi et al. · Language models & NLP
Training data attribution (TDA) methods aim to attribute model outputs back to specific training examples, and the application of these methods to large language model (LLM) outputs could significantly advance model transparency and data curation.
- 2024-10arXiv 2410.15639Can Large Language Models Invent Algorithms to Improve Themselves?: Algorithm Discovery for Recursive Self-Improvement through Reinforcement Learning
Yoichi Ishibashi, Taro Yano, Masafumi Oyamada · Language models & NLP
Large Language Models (LLMs) have achieved remarkable capabilities, yet their improvement methods remain fundamentally constrained by human design. We present Self-Developing, a framework that enables LLMs to autonomously discover, implement, and refine their own improvement algorithms.
- 2024-10arXiv 2410.16090Analysing the Residual Stream of Language Models Under Knowledge Conflicts
Yu Zhao, Xiaotang Du, Giwon Hong et al. · Language models & NLP
Large language models (LLMs) can store a significant amount of factual knowledge in their parameters. However, their parametric knowledge may conflict with the information provided in the context.
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.
