Agreement versus Softmax under Spatial Disambiguation
cs.LG, cs.CV
Submitted: 2026-10-09
Updated: 2026-10-09
Comments: 7 pages, 2 figures. Accepted at the NeurIPS 2026 Workshop on Embodied Spatial Reasoning (non-archival)
License: http://creativecommons.org/licenses/by/4.0/
The gist: ScanRefer Multiple queries require spatial disambiguation: several instances of the target class are present, and the referencer must pick one.
Abstract
ScanRefer Multiple queries require spatial disambiguation: several instances of the target class are present, and the referencer must pick one. When per-query logits were never logged, a common substitute is cross-epoch agreement. On the full ScanRefer validation set (9,508 queries; GeoClassifier with ground-truth proposals), object-level softmax is similarly miscalibrated on Unique and Multiple (ECE 0.068 vs. 0.102; the scene-bootstrap gap interval includes zero). Fixed-final agreement on the same checkpoints reports ECE 0.041 vs. 0.272 (gap +0.231). Modal-vote agreement is 0.045 vs. 0.318. Most Multiple errors are same-class wrong-instance (45.6%); geometric miss (H2) is 59.6%, and wrong-category (H1) is 16.7%. A non-monotone H1 pattern does not appear under the default fixed-final protocol, and we retract it. Logit-free thresholds rank spatially hard queries poorly. If only agreement was logged, calibrate abstention on Multiple and report wrong-instance rates with H1.
Sources
- 3D-LLM: Injecting the 3D World into Large Language Models
- 3D-VCD: Hallucination Mitigation in 3D-LLM Embodied Agents through Visual Contrastive Decoding
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks