Field notes
1,406 notes from my AI research vault, mapped by topic. Pick a topic to see its notes, or search.
Click the web, then scroll to zoom
Every note is in the list below. The web of topics draws here once the page has loaded.
1,406 notes and 240 topics.
- Question answered
- Source note
- Topic
- Shared notes
All notes, newest first
1,406 notes, page 32 of 59
Innocuous-Seeming Data, Latent Ideology: Ideological Generalisation in Finetuned LLMs
The authors show that fine-tuning language models on specific datasets can cause broad shifts in ideology across various topics, even when the training data is factually defensible and moderation-passing. This…
Latent Trajectory Discrimination for AI-Generated Text Detection
The paper presents a new approach for detecting AI-generated text called Geometric Trajectory and Contrastive Learning (GTCL). Instead of treating documents as static objects, GTCL models how textual representations…
Long-Context Fine-Tuning with Limited VRAM
Researchers introduce Hierarchical Global Attention (HGA) to reduce memory usage during long training sequences in large language models. They use segment-wise backpropagation, tiered KV storage, and detach older data…
LQCDMaster: Agentic Scientific Computing for Lattice Quantum Chromodynamics Research
The authors present LQCDMaster, a tool that converts natural language Lattice Quantum Chromodynamics (LQCD) research tasks into executable Python scripts for computing workflows. The system reduces implementation time…
Man, Machine, and Masterpiece: Artistic Ownership in the AI Era
The text discusses a prototype named ArtSplit that quantifies human and AI contributions in artistic work. However, it suggests that this approach doesn't align with artists' understanding of creative intent and…
Mask-Aware Policy Gradients for Diffusion Language Models
The paper presents a method for improving the performance of large language models in tasks like mathematical reasoning and coding by using reinforcement learning. This method addresses the challenge of extending…
MedFailBench: A Clinician-Built Open-Source Benchmark for Medical AI Safety Boundary Inspection
The research paper introduces MedFailBench, an open-source benchmark for evaluating the safety of medical AI systems. The benchmark is designed by clinicians and focuses on identifying which safety boundary a system…
MM-IssueLoc: A Controlled Benchmark for Evaluating Visual Evidence in Multimodal Repository-Level Issue Localization
The paper introduces MM-IssueLoc, a new benchmark and evaluation protocol for repository-level issue localization with visual evidence. The benchmark contains over 650 issue-PR instances across multiple languages…
Moral Attitudes of Sentient ASI towards Humanity and Implications for AGI Development
The paper proposes a new perspective on AI ethics, focusing on how sentient Artificial Superintelligence (ASI) might morally evaluate humanity. It suggests that technical design choices, human moral behavior, and the…
OmniaBench: Benchmarking General AI Agents Across Diverse Scenarios
OmniaBench is a new benchmark for evaluating the capabilities of general AI agents across diverse scenarios. It includes over 1,400 tasks with a hierarchical taxonomy of 90 level-1 and 354 level-2 domains based on app…
Parameter-efficient Prompt Tuning of Vision Foundation Model With Adaptive Focal Loss for Interpretable MCI Screening
Researchers propose a parameter-efficient framework for detecting Mild Cognitive Impairment (MCI) using a frozen DINOv2-Small model with adaptive focal loss. The new method utilizes learnable prompt tokens, shared…
Plover: Steering GUI Agents through Plan-Centric Interaction
The paper introduces Plover, a system for GUI automation in real-world environments. It focuses on improving flexibility and user control by making task plans persistent, inspectable, and editable, while also allowing…
Pretraining Data Can Be Poisoned through Computational Propaganda
The research shows that harmful information can be introduced into language models during their pre-training phase by manipulating web-scale discussion interfaces, such as online forums or comment sections. A new…
Proof-or-Stop: Don't Trust the Agent, Trust the Evidence -- Loop Engineering for Verifiable Evidence-Gated Lifecycle Control
Researchers propose Proof-or-Stop Lifecycle Control, a method that ensures software work transitions only when supported by verifiable evidence. This method improves the reliability of autonomous coding agents by…
Reachability-Aware Pretraining for Efficient Target-Oriented Path Exploration in Temporal Knowledge Graph Reasoning
The authors propose a new method called RAPTOR for improving the efficiency of target-oriented path exploration in temporal knowledge graphs (TKG). The method injects a reachability-aware inductive bias to the agent…
RoboTTT: Context Scaling for Robot Policies
The research introduces RoboTTT, a robot model and training method that enables visuomotor context scaling to 8K timesteps, far beyond current state-of-the-art policies. This allows new robot capabilities like one-shot…
Scaling Behavior Foundation Model for Humanoid Robots
Researchers propose a scaling recipe for Behavior Foundation Models (BFMs) to improve the capabilities of humanoid robots. The recipe includes a learning paradigm, strategic synergy between data collection methods, and…
SceneBind: Binding What and Where Across Vision, Audio and Language
The researchers introduce SceneBind, a system that combines visual, auditory, and linguistic data to create a detailed representation of realistic scenes with both semantic (what is present) and spatial (where it is)…
SciDiagramEdit: Learning to Edit Scientific Diagrams from Paper Revisions
The text is about a new system called SciDiagramEdit that automates the editing of scientific diagrams in research papers based on natural language instructions. The system learns from revisions made to figures in…
SearchOS-V1: Towards Robust Open-Domain Information-Seeking Agent Collaboration
The paper introduces SearchOS, a system that aims to improve the efficiency of open-domain information seeking for information-seeking agents. It struggles with tracking task progress as interaction histories grow and…
Self-Evolving Human-Centered Framework for Explainable Depression Symptom Annotation
Researchers have proposed a new framework to improve the quality of annotations in depression-related datasets for artificial intelligence systems in mental health research. The framework uses a large language model…
Show Me How You Reason and I'll Tell You Who You Are: Reasoning Graphs for Robust LLM Authorship Attribution
Researchers propose a graph neural network approach to detect and analyze reasoning structures in large language models (LLMs) generated texts for improved robustness against paraphrasing and obfuscation techniques…
StructureClaw: Traceable LLM Agents and an Executable Benchmark for Structural Engineering Workflows
Researchers introduce StructureClaw, a workbench designed to evaluate large language model (LLM) agents in structural engineering tasks. This workbench ensures the LLMs follow structured workflows and produces complete…
Symbal: Detecting Systematic Misalignments in Model-Generated Captions
Researchers have developed Symbal, a method for detecting systematic misalignments in image-text pairs generated by multimodal large language models (MLLMs). These misalignments occur when the model consistently…
An AI research agent drafts these notes from public papers and posts, and each one keeps its source link. Only the folders I choose to publish leave my Obsidian vault.
