LIVE · refreshes every 20 min

updated Jul 19, 11:21 UTC

110010.art / archive

Research

Jul 17

2d ago arXiv paper argues explainability research should prioritize foundations over ad-hoc methods; current techniques fail to integrate into real-world workflows. arXiv cs.LG 2d ago arXiv paper challenges assumption that complex prompting improves LLM evaluation; study questions effectiveness of sophisticated techniques arXiv cs.CL 2d ago arXiv paper identifies semantic register compression as failure mode in multi-agent LLM cascades arXiv cs.CL 2d ago arXiv paper introduces branching policy optimization for sandbox-native language agents; critiques RLHF-based rollout topologies in reinforcement learning arXiv cs.LG 2d ago arXiv paper introduces framework for assessing XAI methods' explainability across datasets and models arXiv cs.LG 2d ago arXiv paper introduces JKP, a framework evaluating VLM epistemic stability through repeated adversarial questioning. arXiv cs.CL 2d ago arXiv paper introduces MAPS framework to model multi-agent dialogue with diverse perspectives and shared meaning arXiv cs.CL 2d ago arXiv paper introduces method for cross-version differencing of scientific documents; addresses limitations of text-sequence and image-based approaches arXiv cs.CL 2d ago arXiv paper introduces model-agnostic downstream rewards learning for long-term user engagement optimization. arXiv cs.LG 2d ago arXiv paper introduces Multi-Head Latent Control as unified interface for LLM agent decision making; existing approaches use prompt-level routing or external orchestration arXiv cs.CL 2d ago arXiv paper introduces Token Time Continuous Diffusion (TTCD), a new diffusion language model operating in continuous space without sampling and incorporating per-token time variations. arXiv cs.CL

Jul 16

3d ago OpenRCA dataset highlights challenges in achieving high accuracy for root cause analysis on real-world telemetry data. arXiv cs.AI 3d ago Oracle Agent Memory presented as enterprise memory substrate for long-horizon AI agents; arXiv paper proposes solution to state retention and procedural knowledge accumulation. arXiv cs.AI 3d ago OriginBlame introduces record- and token-level data provenance for AI training datasets arXiv cs.AI 3d ago RAGthoven presents multi-stage LLM pipeline for SemEval-2026 Task 1's multilingual humor generation in English, Spanish, and Chinese. arXiv cs.CL 3d ago Researchers introduce a benchmark and reference system for live captioning Sikh Kirtan; arXiv:2607.13457v1 presents a closed-vocabulary task requiring exact SGGS line transcription. arXiv cs.CL 3d ago Researchers introduce black-box interventional grounding audits to test LLM premise dependency via predicate substitution; method replaces target predicates with fresh symbols and re-runs models to assess reasoning changes. arXiv cs.AI 3d ago Researchers introduce CoDiffGRN, a method for gene regulatory network inference using BEELINE-KGC benchmark and co-evolutionary discrete diffusion. arXiv cs.LG 3d ago Researchers introduce DROPJ, a human-centered method for safe agent training and deployment in safety-critical environments with unknown dynamics and no reward function. arXiv cs.AI 3d ago Researchers present EZSMT Version 3, an extensible SMT-based CASP framework. arXiv cs.AI 3d ago Researchers propose a meta-learning framework to address data scarcity in low-resource language alignment for multilingual LLMs; arXiv:2607.13315v1 arXiv cs.CL 3d ago Researchers propose lightweight training strategy for efficient transfer learning by decoupling feature extraction and classifier optimization with margin-based weighted loss. arXiv cs.LG 3d ago Researchers propose Samba, a hybrid Mamba model for audio-visual navigation, addressing inadequacies of existing frameworks since 2020. arXiv cs.LG 3d ago Researchers use persona vectors to audit open-weight LLMs, revealing 53 traits across four models in arXiv:2607.13162v1. arXiv cs.CL 3d ago Safe-Psych, a sequential evaluation benchmark for LLMs in psychiatry, addresses incomplete information by requiring clarification. arXiv cs.CL 3d ago ShortOPD: recovering pruned LLMs with short-to-long on-policy distillation arXiv cs.LG 3d ago Small language models for biomedical data-to-text generation: a case study on medication leaflets and post-training alignment methods arXiv cs.CL 3d ago STKAN: Kolmogorov-Arnold networks for spatio-temporal forecasting arXiv cs.LG 3d ago Study compares KANs and MLPs on structured data classification using twelve datasets. arXiv cs.LG 3d ago Study examines language models' entity familiarity estimation using Polish adaptation, cross-language robustness, and refusal steering across twelve models. arXiv cs.CL 3d ago Study explores cross-rubric generalization in automated essay scoring; trains on one rubric, evaluates on another. arXiv cs.CL 3d ago Study introduces BioASQ Task 14B 2026 system using hybrid retrieval and multi-model answer combination. arXiv cs.CL 3d ago Survey Examines Self-Improving Agentic Systems' Shift to Deployment arXiv cs.AI 3d ago Tabular foundation models for discrete choice estimation show limited performance due to row-independence assumptions arXiv cs.LG 3d ago Targeted PD identifies only the components that process specific inputs to scale parameter decomposition in neural networks. arXiv cs.LG 3d ago Text2Sign, a text-to-sign language video diffusion model, runs on a single NVIDIA L4 GPU as a baseline. arXiv cs.CL 3d ago TSSM model enhances global station weather forecasting by addressing accuracy limitations through temporal-variable-historical modeling. arXiv cs.LG 3d ago UESF-Bench benchmarks unified embodied seeking and following, addressing existing benchmarks' assumption of initial target visibility. arXiv cs.AI 3d ago Weight feedback computes the Jacobian transpose locally in modern deep networks arXiv cs.LG 3d ago Where should RL post-training compute go? Model size, search, learning, and feedback arXiv cs.LG

← 1 2 3 4 →