Skip to main content
Search

Texas A&M University

18 related articles

Subscribe to updates
  1. bioRxiv

    Protein language model steering drops the global mean: PULSE pairs similar backgrounds, with the largest predicted-score gain on toxicity

    Protein language models can be steered toward a target property without changing their weights, but existing activation steering builds the direction from the difference between the global means of property-positive and property-negative proteins, ignoring that the useful direction shifts with sequence background. PULSE first matches positive and negative proteins by representation similarity, then aggregates similarity-weighted, unit-normalized differences between matched pairs into a steering vector. On solubility, immunogenicity and toxicity optimization, PULSE beats global centroid steering in both final predicted property scores and optimization success rates, with the largest score gain on toxicity and the largest success gain on immunogenicity.

  2. arXiv

    World-model test-time adaptation stops fitting prediction alone: OpTTA adds an optimistic planning term and beats AdaJEPA on final success in 14 of 15 shifted settings

    Pretrained latent world models are usually frozen at test time, and distribution shift can degrade planning with them; updating only on prediction loss fits the outcomes of the agent's own actions and can stay wrong about actions that would reach the goal. OpTTA adds the planner's goal-reaching cost on candidate imagined rollouts to the prediction loss, biasing adaptation toward model parameters that are optimistic about reaching the goal. Across three JEPA world models and 15 shifted settings, adapting only the predictor raises final success over the frozen model by 13.5 percentage points on average, versus 6.9 for prediction-only AdaJEPA; with the encoder also adapted, the gains are 13.5 versus 7.5 points.

  3. bioRxiv

    CARTnet treats known epitopes as templates: internal AUPRC rises from 0.225 to 0.403, and it also leads on an external set

    B-cell epitope prediction has been constrained by scarce experimentally annotated data, and CARTnet feeds sequence, structure, physicochemical features and an optional epitope template into an attention network with contact bias. On a 61-target internal test set its no-template version reached an AUPRC of 0.243 against 0.220 for DiscoTope-3, and on a 52-target external set it reached 0.267 against 0.215 for BepiPred-3. The template-guided variant rose from 0.225 to 0.403 AUPRC on the internal set with a single template residue, though the gain saturated as more templates were supplied.

  4. arXiv

    CyberArena puts AI agents in live attack-and-defense combat, and two frontier models pull ahead of the field

    CyberArena turns cybersecurity evaluation from static targets into multi-round combat: two agents in isolated sandboxes attack each other's services while patching their own, with objectives rotating every round. Across 40 challenges and 466 games, GPT-5.6 Sol and Claude Opus 5 led 12 frontier agents, and on 30 challenges all three compared models solved solo, round-three capture rates fell to 52%, 37% and 35%.

  5. arXiv

    Wildfire forecasts move to a 12-hour clock: FireGraphNet's directed graph lifts near-front ranking 1.7–1.9 times a distance rule

    Wildfire decisions are made in hours, yet most learned spread models predict daily on targets that include already burned area. FireGraphNet recasts roughly 12-hour growth as node classification on directed H3 hexagonal graphs: candidates are unburned cells within five rings of the front, edges carry wind alignment, elevation difference, fuel continuity and barriers, and a graph neural network passes messages along those edges. On five sealed California test fires the five variants reach average precision 1.7–1.9 times the strongest distance baseline, and the GINE variant keeps a 1.62-fold advantage out of state.

  6. bioRxiv

    No fixed target for tongue motor networks: individualized AI neuromodulation lifts target-state occupancy from 82% to 97%

    The individualized AI-driven fMRI neuromodulation platform iNM abandons a single fixed set of brain targets and instead builds a neurofunctional twin from each participant's own anatomy and function, reinforcing the tongue sensorimotor network in a closed loop at 2-second resolution. In healthy participants, an unsupervised hidden Markov model showed target-state occupancy rising from 82% under control to 97%, swallow-related interference falling from 18% to 3%, and connectivity among motor, sensory and attention networks strengthening by 53% to 74%, with effects persisting after modulation stopped.

  7. bioRxiv

    SVC treats each gene's intracellular spread as an image patch: other genes in the same cell supply the context for predicting where unmeasured transcripts sit

    Subcellular-resolution spatial transcriptomics can locate transcripts inside cells, but earlier methods targeted single analysis tasks and lacked a unified framework for characterizing and predicting gene localization in a cell-specific, context-aware way. SVC represents each gene's within-cell spatial expression as an image patch and uses other genes in the same cell as context, learning cross-gene spatial dependencies through self-supervised masking. On seqFISH+ mouse embryonic fibroblast data, SVC beat four image-based baselines on all six metrics for predicting subcellular patterns of unmeasured genes, with median gains of 5.2%–21.0% over the more challenging baseline pair.

  8. arXiv

    Robot disassembly learns from its own rollouts: PACER turns autonomous traces and human corrections into training weights, reaching 87/100 motherboard handoffs

    In electronic-waste disassembly, demonstrations for vision-language-action (VLA) policies are scarce, and supervised fine-tuning alone struggles to cover the state shifts a deployed policy meets. PACER compiles autonomous rollouts, operator corrections and supplementary clean demonstrations into typed action-chunk rows, then weights each row by progress toward the declared target and by the trace's source and role, while keeping the native supervised action loss. On held-out hardware for UR5e motherboard disassembly across five components, the selected policy completed 87/100 approach-and-handoff trials, against 54/100 for uniformly weighted post-SFT and 79/100 for fixed-geometry reweighting.

  9. arXiv

    Structured knowledge retrieval drops hand-written query-rule pairs: a routing agent and its retriever train in alternation

    Structured knowledge retrieval no longer needs predefined query-rule pairs: a routing agent picks a reasoning rule per query, its retriever ranks evidence under that rule's constraints, and the two train in alternation, each supervising the other.

  10. arXiv

    Robot rewards stop chasing lighting and camera angle: task-relevant dynamics lift MetaWorld success to 94.3%

    DiTaR builds video rewards on task-relevant latent dynamics, reaching 94.3% final success on MetaWorld-v3 versus 65.2% for the strongest video-reward baseline ReWiND, and raising real-robot offline RL success from 43.3% to 83.3%.