Deep Value Networks Learn to Evaluate and Iteratively Refine Structured Outputs

2017-03-13 · ICML 2017 8 · Michael Gygli, Mohammad Norouzi, Anelia Angelova

We approach structured output prediction by optimizing a deep value network (DVN) to precisely estimate the task loss on different output configurations for a given input. Once the model is trained, we perform inference by gradient descent on the continuous relaxations of the output variables to find outputs with promising scores from the value network. When applied to image segmentation, the value network takes an image and a segmentation mask as inputs and predicts a scalar estimating the intersection over union between the input and ground truth masks. For multi-label classification, the DVN's objective is to correctly predict the F1 score for any potential label configuration. The DVN framework achieves the state-of-the-art results on multi-label prediction and image segmentation benchmarks.

📄 PDF Abstract BibTeX arXiv:1703.04363

Code (1)

gyglim/dvn 공식 구현 pytorch

Tasks

General Classification Image Segmentation Multi-Label Classification MUlTI-LABEL-ClASSIFICATION Segmentation Semantic Segmentation

Automatic Prompt Optimization for Dataset-Level Feature Discovery

2026-01-20 · Adrian Cosma, Oleg Szehr, David Kletz, Alessandro Antonucci 외 arxiv

Feature extraction from unstructured text is a critical step in many downstream classification pipelines, yet current approaches largely rely on hand-crafted prompts or fixed feature schemas. We formulate feature discove…

GRADRAG: Cross-Component Prompt Adaptation for Coordinated Multi-Agent RAG

2026-07-23 · Paolo Pedinotti, Enrico Santus arxiv

Retrieval-Augmented Generation (RAG) systems increasingly employ multiple LLM agents. Yet, most prior work optimizes components in isolation rather than coordinating improvements across the pipeline. We introduce GRADRAG…

BDI-Kit Demo: A Toolkit for Programmable and Conversational Data Harmonization

2026-04-07 · Roque Lopez, Yurong Liu, Christos Koutras, Juliana Freire arxiv

Data harmonization remains a major bottleneck for integrative analysis due to heterogeneity in schemas, value representations, and domain-specific conventions. BDI-Kit provides an extensible toolkit for schema and value …

Automated Self-Refinement and Self-Correction for LLM-based Product Attribute Value Extraction

2025-01-02 · Alexander Brinkmann, Christian Bizer

Structured product data, in the form of attribute-value pairs, is essential for e-commerce platforms to support features such as faceted product search and attribute-based product comparison. However, vendors often provi…

Attribute Attribute Value Extraction Code Generation In-Context Learning+2

Test-Driven Agentic Framework for Reliable Robot Controller

2026-02-28 · Shivanshu Tripathi, Reza Akbarian Bafghi, Maziar Raissi arxiv

In this work, we present a test-driven, agentic framework for synthesizing a deployable low-level robot controller for navigation tasks. Given a 2D map with an image of an ultrasonic sensor-based robot, or a 3D robotic s…