Fine-tuning

Timeline

1. Oct 8, 2026 · Research paper · 1 source
   SpatialHarness: Test-Time Spatial Scaffolding for Fine Robotic Manipulation
   We introduce SpatialHarness, a test-time embodied harness that provides test-time spatial scaffolding for fine robotic manipulation without policy fine-tuning or changes to the physical sensing setup.
2. Oct 8, 2026 · Research paper · 1 source
   Rounding in Preconditioner Space: Redesigning 4-bit AdamW Optimizer-State Quantization
   Quantizing AdamW's optimizer states reduces persistent storage, but quantization errors propagate through the moment recurrences and perturb subsequent adaptive updates.
3. Oct 8, 2026 · Research paper · 1 source
   VioLA: Learning Generalist Humanoid Control Policies from Human Data
   We introduce VioLA, a generalist humanoid policy that predicts body and hand motion latents instead of joint commands.
4. Oct 8, 2026 · Research paper · 1 source
   Predicting Alignment Generalization with Value Representations
   In this paper, we establish the task of alignment generalization prediction, i.e., predicting how fine-tuning a model to follow a given value changes its behavior across a wide range of held-out values.
5. Oct 8, 2026 · Research paper · 2 sources
   SpaceCast-Bench: Evaluating Predictive Spatial Reasoning in Vision-Language Models
   Existing spatial reasoning benchmarks mainly test spatial perception: reading off relations already visible in the input.
6. Oct 8, 2026 · Research paper · 1 source
   Learning Kilometer-Scale Weather Prediction with Global-Regional Alignment
   We propose ScaleCast, a regional forecasting framework that addresses these challenges through Global-Regional Alignment.
7. Oct 8, 2026 · Research paper · 1 source
   ARC: A Reasoning Recipe for Robot Foundation Models
   We show that there exists an effective and efficient complementary approach: the right reasoning recipe can substantially improve the zero-shot task performance of existing state-of-the-art RFMs. We refer to this recipe as ARC.
8. Oct 8, 2026 · Research paper · 1 source
   Overcoming Prior Barriers: Supervised Fine-Tuning under Long-Tail Distribution
   Supervised fine-tuning (SFT) adapts pretrained large language models (LLMs) to downstream tasks, but the required concepts can receive substantially different levels of pretrained support.
9. Oct 8, 2026 · Research paper · 1 source
   Ambient Discrete Diffusion: Using the Wrong Data at the Right Time for Data Efficient Learning
   We introduce RefineMix, a framework for training discrete diffusion models under severe data scarcity, a common constraint in scientific applications.
10. Oct 8, 2026 · Research paper · 1 source
    Prior or Feedback? What an LLM Uses When Adapting Neural Operators
    Do LLM scientific agents rely only on their initial task context, or do they adapt their decisions in response to experimental feedback?
11. Oct 8, 2026 · Research paper · 1 source
    Controllable Exaggeration for Generative Motion Models via Training-Time Adaptation and Inference-Time Guidance
    Recent motion generative models have demonstrated strong capabilities in synthesizing physically plausible character motion, but often overlook established animation principles used by professional animators to ground and design their animation work.
12. Oct 8, 2026 · Research paper · 1 source
    HarnessSQL: Harness-Native Training for SQL Agents in Realistic Database Environments
    To bridge this gap, we propose HarnessSQL, a harness-native post-training framework that preserves the full interaction structure throughout both supervised fine-tuning and reinforcement learning.
13. Oct 8, 2026 · Research paper · 2 sources
    VibeEdit: Image Editing with Canvas Instructions
    We introduce a new image editing interface that lets users place spatial marks and optional short notes directly on the image.
14. Oct 8, 2026 · Research paper · 1 source
    Recursive Self-Improvement through Multi-Agent Self-Supervision
    To address this, we propose Multi-Agent Self-Supervision (MASS), an RSI method that alternates between evolutionary workflow optimization and supervised fine-tuning on self-generated trajectories.
15. Oct 8, 2026 · Research paper · 1 source
    Is Real-World Training Data Necessary for Generalist Graph Anomaly Detection?
    Generalist graph anomaly detection (GAD) aims to build a foundation model that detects anomalies on arbitrary unseen graphs without retraining or fine-tuning.
16. Oct 8, 2026 · Research paper · 1 source
    An Investigation of Model Coherence: Narrow Finetunes Contradict Themselves Under Resampling
    A large body of research measures model coherence based on output variance without adequately considering competing causes.
17. Oct 8, 2026 · Research paper · 2 sources
    SuperNav: An Agentic Navigation System for Any Task in Any Scene
    General-purpose service robots need navigation systems that can handle diverse human requests in unfamiliar environments, combining task generality with scene generality.
18. Oct 8, 2026 · Research paper · 1 source
    When Should Agents Think? Adaptive Reasoning via Cross-Turn Estimation
    Based on this observation, we propose Reasoning Adaptation through Cross-Turn Estimation (RACE), a training approach for adaptive agent reasoning.
19. Oct 8, 2026 · Research paper · 1 source
    Do Not Train Away Uncertainty: Early Uncertainty Anchored Calibration
    EUA-Cal introduces early prediction regularization to preserve early predictive uncertainty and prototype structure regularization to exploit uncertainty reflected in the early feature space, jointly mitigating overconfidence.
20. Oct 8, 2026 · Research paper · 1 source
    Natural Language to First-Order Logic LLM-based Autoformalization
    This paper addresses this gap: we first provide a principled definition for the FOL-autoformalization task by distinguishing Ontology Extraction from Logical Translation, showing how their conflation obscures (cross-study) evaluation; we review existing datasets, evaluation metrics, and LLM-based methods, including fine-tuning, prompting, and verification-based refinement; we identify open challenges in benchmarking, semantic evaluation, ontology-aware methods, and end-to-end applications.
21. Oct 8, 2026 · Research paper · 1 source
    PulseBound: Future-Beat State Forecasting Under an Explicit Information Boundary
    We introduce PulseBound, a PPG representation learner combining physiologically structured future-beat prediction with an explicit stored-window information boundary.
22. Oct 8, 2026 · Research paper · 1 source
    Test-Time Compute for Tabular Foundation Models: Mechanisms, Gains, and Limits
    Which forms of test-time compute improve the predictions of strong pretrained tabular foundation models (TFMs)?
23. Oct 8, 2026 · Research paper · 1 source
    Project Greenhouse: Progress Toward Fully Open and Sovereign Agentic Search
    Project Greenhouse represents our exploration of a simple thesis: We believe that it is possible to build fully open and sovereign models for agentic search with only modest computational resources.
24. Oct 8, 2026 · Research paper · 1 source
    From Pixels to Structure: Lightweight Vision-Language Models for Document OCR and Structured JSON Extraction
    We present a comparative study of eight open-source lightweight VLMs (up to 7B parameters) for Optical Character Recognition (OCR)-to-structure across three university heritage collections.
25. Oct 8, 2026 · Research paper · 1 source
    Seek-and-View Reasoning for Multi-View Spatial Understanding
    To realize this approach, we propose Vantage, a training-free model-agnostic reasoning framework that pairs a VLM with a 3D foundation model: a viewpoint-grounded reasoning stage for question analysis and view planning, followed by a geometry-grounded evidence augmentation stage to effectively synthesize and incorporate visual evidence into the final reasoning.
26. Oct 8, 2026 · Research paper · 1 source
    Easy to anticipate, hard to compute: boundary dependence finds the computed outputs that entropy patching misses
    Byte-level language models such as the Byte Latent Transformer (BLT) group bytes into patches and run their large global model once per patch.
27. Oct 8, 2026 · Research paper · 1 source
    MAP4CS: A Multi-dimensional Data Pruning Framework for Efficient Code Retriever Fine-tuning
    To address these challenges, we propose MAP4CS (Multi-dimensional Awareness Pruning for Code Search), an adaptive data pruning framework.
28. Oct 8, 2026 · Research paper · 1 source
    From Solo to Ensemble: A Hierarchical Framework for Composable Multi-Agent Human-Object Interaction
    We propose a hierarchical framework that converts a single-agent HOI policy into a reusable Object-oriented Motion Skill.
29. Oct 8, 2026 · Research paper · 1 source
    Autoregressive Retriever: Improving Query Understanding from Item Feedback for Universal Multimodal Retrieval
    We introduce the AutoRegressive Retriever (ARR), a multimodal retrieval model that learns both to select informative items and to use their content to refine subsequent retrieval.
30. Oct 8, 2026 · Research paper · 1 source
    YOCO: You Only Calibrate Once! Fast Mocap Calibration for Dexterous Teleoperation
    We present YOCO, a fast few-shot, fine-tuning-free calibration framework that corrects biased hand-pose streams from a small set of paired raw and target poses.

Often appears with