LA
Laurence Aitchison
cs.LGstat.MLcs.AIcs.CLq-bio.NCcs.CVcs.ROModels, Neurologicalneurosciencephysics.ao-ph
On Valency
published · living versionsW_5txyfx4z·v1 · currentpublished
Preprints & journals
86 papers in the corpus · 2014–2026Unifying error and reward action learning: a cerebello-basal ganglia theory10.64898/2026.08.22.745751v1 · Garibbo, M., Filipe, C., Aitchison, L. et al.2026 · 0 citationsbioRxiv
Le Critique: Privileged Value Functions for LLM Reinforcement Learning2608.16739v1 · Siddarth Venkatraman, Matthieu Dinot, Laurence Aitchison2026 · 0 citationsarXiv
Shieldstral2607.25857v2 · Antonia Calvi, Avinash Sooriyarachchi, Giada Pistilli et al.2026 · 0 citationsarXiv
Robostral Navigate2607.20785v3 · Abdelaziz Bounhar, Abhijeet Somani, Aditi Kabra et al.2026 · 0 citationsarXiv
Reasoning Arena: Trace Tournaments When Verifiable Rewards Fall Short2606.09380v1 · Han Zhou, Adam X. Yang, Laurence Aitchison et al.2026 · 0 citationsarXiv
Inverse-Free Sparse Variational Gaussian Processes2604.00697v1 · Stefano Cortinovis, Laurence Aitchison, Stefanos Eleftheriadis et al.2026 · 0 citationsarXiv
Voxtral TTS2603.25551v2 · Mistral-AI: Alexander H. Liu, Alexis Tacnet, Andy Ehrenberg et al.2026 · 0 citationsarXiv
Learning Generation Orders for Masked Discrete Diffusion Models via Variational Inference2602.23968v1 · David Fox, Sam Bowyer, Song Liu et al.2026 · 0 citationsarXiv
Rodent-Bench2602.18540v1 · Thomas Heap, Laurence Aitchison, Emma Cahill et al.2026 · 0 citationsarXiv
Voxtral Realtime2602.11298v3 · Mistral-AI: Alexander H. Liu, Andy Ehrenberg, Andy Lo et al.2026 · 0 citationsarXiv
Ministral 32601.08584v1 · Alexander H. Liu, Kartik Khandelwal, Sandeep Subramanian et al.2026 · 5 citationsarXiv
Controlling changes to attention logits2511.21377v1 · Ben Anson, Laurence Aitchison2025 · 0 citationsarXiv
Learning to Skip the Middle Layers of Transformers2506.21103v1 · Tim Lawson, Laurence Aitchison2025 · 0 citationsarXiv
Improving LLM-Generated Code Quality with GRPO2506.02211v1 · Maxime Robeyns, Laurence Aitchison2025 · 0 citationsarXiv
Scale-invariant Attention2505.17083v2 · Ben Anson, Xi Wang, Laurence Aitchison2025 · 0 citationsarXiv
A Self-Improving Coding Agent2504.15228v2 · Maxime Robeyns, Martin Szummer, Laurence Aitchison2025 · 1 citationarXiv
Massively Parallel Expectation Maximization For Approximate Posteriors2503.08264v1 · Thomas Heap, Sam Bowyer, Laurence Aitchison2025 · 0 citationsarXiv
Position: Don't Use the CLT in LLM Evals With Fewer Than a Few Hundred Datapoints2503.01747v3 · Sam Bowyer, Laurence Aitchison, Desi R. Ivanova2025 · 0 citationsarXiv
Jacobian Sparse Autoencoders: Sparsify Computations, Not Just Activations2502.18147v2 · Lucy Farnik, Tim Lawson, Conor Houghton et al.2025 · 0 citationsarXiv
Function-Space Learning Rates2502.17405v2 · Edward Milsom, Ben Anson, Laurence Aitchison2025 · 0 citationsarXiv
Automated Interpretability Metrics Do Not Distinguish Trained and Random Transformers2501.17727v2 · Thomas Heap, Tim Lawson, Lucy Farnik et al.2025 · 0 citationsarXiv
Relating Human Error-Based Learning to Modern Deep RL Algorithms.39383023 · Garibbo, Michele, Ludwig, Casimir J H, Lepora, Nathan F et al.2024 · 0 citationsNeural computation. 2024;37(1):128-159
Why you don't overfit, and don't need Bayes if you only train for one epoch2411.14478v1 · Laurence Aitchison2024 · 0 citationsarXiv
Human-inspired Perspectives: A Survey on AI Long-term Memory2411.00489v2 · Zihong He, Weizhe Lin, Hao Zheng et al.2024 · 3 citationsarXiv
Stochastic Kernel Regularisation Improves Generalisation in Deep Kernel Machines2410.06171v1 · Edward Milsom, Ben Anson, Laurence Aitchison2024 · 0 citationsarXiv
Residual Stream Analysis with Multi-Layer SAEs2409.04185v3 · Tim Lawson, Lucy Farnik, Conor Houghton et al.2024 · 0 citationsarXiv
Signatures of Bayesian inference emerge from energy-efficient synapses.39106188 · Malkin, James, O'Donnell, Cian, Houghton, Conor J et al.2024 · 3 citationseLife. 2024;12
Machine learning emulation of precipitation from km-scale UK regional climate simulations using a diffusion model2407.14158v3 · Henry Addison, Elizabeth Kendon, Suman Ravuri et al.2024 · 7 citationsarXiv
Questionable practices in machine learning2407.12220v2 · Gavin Leech, Juan J. Vazquez, Niclas Kupper et al.2024 · 5 citationsarXiv
Using Neural Networks for Data Cleaning in Weather Datasets2406.15027v1 · Jack R. P. Hanslope, Laurence Aitchison2024 · 0 citationsarXiv
Instruction Tuning With Loss Over Instructions2405.14394v2 · Zhengyan Shi, Adam X. Yang, Bin Wu et al.2024 · 4 citationsarXiv
How to set AdamW's weight decay as you scale model and dataset size2405.13698v3 · Xi Wang, Laurence Aitchison2024 · 1 citationarXiv
Snap-it, Tap-it, Splat-it: Tactile-Informed 3D Gaussian Splatting for Reconstructing Challenging Surfaces2403.20275v1 · Mauro Comi, Alessio Tonioni, Max Yang et al.2024 · 6 citationsarXiv
Batch size invariant Adam2402.18824v1 · Xi Wang, Laurence Aitchison2024 · 0 citationsarXiv
Bayesian Reward Models for LLM Alignment2402.13210v2 · Adam X. Yang, Maxime Robeyns, Thomas Coste et al.2024 · 2 citationsarXiv
Flexible Infinite-Width Graph Convolutional Neural Networks2402.06525v2 · Ben Anson, Edward Milsom, Laurence Aitchison2024 · 0 citationsarXiv
Position: Bayesian Deep Learning is Needed in the Age of Large-Scale AI2402.00809v5 · Theodore Papamarkou, Maria Skoularidou, Konstantina Palla et al.2024 · 7 citationsarXiv
TouchSDF: A DeepSDF Approach for 3D Shape Reconstruction using Vision-Based Tactile Sensing2311.12602v1 · Mauro Comi, Yijiong Lin, Alex Church et al.2023 · 27 citationsarXiv
How to derive skill from the Fractions Skill Score2311.11985v3 · Bobby Antonio, Laurence Aitchison2023 · 5 citationsarXiv
Using Autodiff to Estimate Posterior Moments, Marginals and Samples2310.17374v2 · Sam Bowyer, Thomas Heap, Laurence Aitchison2023 · 0 citationsarXiv
LoRA ensembles for large language model fine-tuning2310.00035v2 · Xi Wang, Laurence Aitchison, Maja Rudolph2023 · 9 citationsarXiv
Convolutional Deep Kernel Machines2309.09814v3 · Edward Milsom, Ben Anson, Laurence Aitchison2023 · 0 citationsarXiv
Signatures of Bayesian inference emerge from energy efficient synapses2309.03194v4 · James Malkin, Cian O'Donnell, Conor Houghton et al.2023 · 4 citationsarXiv
Bayesian Low-rank Adaptation for Large Language Models2308.13111v5 · Adam X. Yang, Maxime Robeyns, Xi Wang et al.2023 · 6 citationsarXiv
An Improved Variational Approximate Posterior for the Deep Wishart Process2305.14454v1 · Sebastian Ober, Ben Anson, Edward Milsom et al.2023 · 1 citationarXiv
Massively Parallel Reweighted Wake-Sleep2305.11022v1 · Thomas Heap, Gavin Leech, Laurence Aitchison2023 · 0 citationsarXiv
Taylor TD-learning2302.14182v2 · Michele Garibbo, Maxime Robeyns, Laurence Aitchison2023 · 0 citationsarXiv
MONGOOSE: Path-wise Smooth Bayesian Optimisation via Meta-learning2302.11533v2 · Adam X. Yang, Laurence Aitchison, Henry B. Moss2023 · 2 citationsarXiv
Decision trees compensate for model misspecification2302.04081v1 · Hugh Panton, Gavin Leech, Laurence Aitchison2023 · 1 citationarXiv
Imitating careful experts to avoid catastrophic events2302.01193v1 · Jack R. P. Hanslope, Laurence Aitchison2023 · 0 citationsarXiv
Career total: 137 works. 86 are in this corpus.Showing the 50 most recent.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.