Influence functions enable principled data selection for large language models (e.g., identifying influential training samples for model editing, debugging, or efficient fine-tuning). However, scaling IFs to LLMs is hindered by the intractability of billion-scale Hessian matrices and the prohibitive memory overhead of...
Li-Song He, Yun-Xuan Jiang, Hai-Shan Ye et al.· 0 citations
We introduce Anchor-and-Harvest Curriculum (AHC), a domain-informed, rule-based training system for Craftax-Full. AHC combines achievement-oriented tasks, harvested initial-state progress, difficulty adaptation, bounded selection, and original-task practice. It uses no LLM inference and keeps all 43 actions available....
Jiayu Zhu, Luyuan Li, Yuyuan Shi et al.· 0 citations
A deception detector is only as useful as the trust its alarms deserve. Like a polygraph, detectors that read a language model's internal activations can catch deceptive responses that look honest. Yet current detectors depend on which layers they read, weaken on broader tests, and do not state how certain their alarms...
Misclassification is a critical problem in many machine learning applications. Since even the classifier models with high accuracy (e.g., > 95%) still introduce some misclassification error, it may not be possible to rely on the output of a classifier. In this paper, we introduce trustable learning, which prompts the l...
Truong X. Tran, Ramazan S. Aygun· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
A training-free PTQ framework for ARVG, consisting of Tail-Aware Bounded-init Grid Refinement (T-BGR) and Coarse-to-Fine Activation Precision Scheduling (CAPS), which achieves near full-precision image generation quality at an effective W4A5.6 activation precision.
Xin-Meng Liu, Xiang-Long Yan, Kai-Cheng Yang et al.· 0 citations
We present Co-DSL, a framework in which users and LLMs co-design reusable spatial concepts for layout generation. In the first stage, our system allows users to create a library of personalized operations. To add a concept, a user provides a description, which the LLM converts into an initial implementation. The user c...
R. Kenny Jones, Jiajun Wu, Maneesh Agrawala· 0 citations
The accuracy of detecting protein crystals for fluorescence microscopy images is very critical for high throughput and automated systems. Although the trace fluorescent labeling method could highlight protein crystals, reflection and emission from the fluorescence dye is not always due to crystal regions. Therefore, th...
Truong X. Tran, Marc L. Pusey, Ramazan S. Aygun· 0 citations
Large Language Models (LLMs) have made substantial progress in multilingual question answering, yet their ability to understand low-resource sociolinguistic contexts remains under-explored. We study North Korea as a unique testbed where decades of division have produced distinct differences from South Korea in vocabula...
Kiwoong Park, Seyoung Song, Donghyun Woo et al.· 0 citations
Depth estimation and Gaussian reconstruction share a need for multi-scale feature processing, yet use different prediction interfaces and training objectives. Whether a decoder's internal computation can be optimized once and reused across these systems remains a structural design question. Search over primitive config...
Joint-embedding predictive architectures (JEPAs) learn world models directly in latent space, but unconstrained autoregressive predictors may contract the latent distribution over long rollouts. We introduce Symplectic Hamiltonian JEPA (LeHami), which replaces the standard predictor with an action-conditioned Hamiltoni...
We introduce Sequent-Prover, a family of verifier-using language-model agents, and the post-training framework that produces them. Agents reason in an executable SMT-LIB scratchpad: they write complete SMT-LIB programs, query Z3 across turns, inspect solver feedback, repair formalizations, and terminate with an explici...
Debargha Ganguly, Wei-Cong Chen, Mohsen Hariri et al.· 4 citations
Self-evolving mechanisms let vision-language model (VLM) agents continually adapt through evolving memory, but they also allow malicious behavior to emerge only after deployment. We introduce a new attack surface, the self-evolving backdoor, whose payload stays dormant in the initial agent and is progressively activate...
Chuan Zou, Yang Xiang, Zexin Li et al.· 0 citations
