Research Paper Cockpit

Daily Digest - 2026-11-13

Papers first seen in this daily snapshot.

Daily Archives

Quick jump into generated daily digests.

Research Workflow

Latest digest: 2026-11-17.

Papers

59 visible entries

arxiv Score 30.0

Maximizing Capacity Utilization of LoRA Adapters for Mitigating the Learning-Forgetting Trade-off

2026-11-12 · Tomohiro Sawada, Kartik Goyal

Research Track A · General AI

Low-Rank Adaptation (LoRA) often underperforms full fine-tuning, especially in heterogeneous and continual-learning settings in which substantial forgetting occurs. In general, it suffers from an undesirable learning-forgetting trade-off across diverse fine-tuning settings. We identify a potential contributor: LoRA ada…

Review
pending
Role
unreviewed
Read
now
arxiv Score 23.5

Reasoning Supervision and Capability Retention in LoRA Domain Adaptation

2026-11-12 · Saleh Ashkboos, Dhruv Desai, Lavinia Ghita

Research Track A · General AI

Fine-tuning foundation models on financial data has a wide range of applications, including identifying event types from news, monitoring risk and compliance language, and answering questions over filings and earnings calls. In this setting, the main challenge is to improve task performance without degrading the genera…

Review
pending
Role
unreviewed
Read
now
huggingface Score 22.4

Agent-R^2: Scaling Agents for Realistic Productivity Work via Real-Source Data Synthesis

2026-11-05 · Teng Pan, Zixuan Wang, Yuchen Yan, Qifei Wu, Yao Chen, Yingqi Qu, Ruiqing Zhang, Yongliang Shen

General AI

Large language models have made rapid progress in agentic reasoning and tool use, raising expectations for agents that can take over realistic productivity work from humans. However, current agents still struggle to finish such tasks, where success requires following long-horizon professional workflows, understanding h…

Review
pending
Role
unreviewed
Read
now
huggingface Score 21.4

HiVe-Mem: Interleaving Horizontal Memory Expansion and Vertical Evidence Composition for Multimodal Agents

2026-11-09 · Haozhen Zhang, Bohan Liu, Jianzhu Bao, Qingyuan Liu, Haodong Yue, Xiao Li, Zi Pong Lim, Wenya Wang

General AI

As multimodal agents accumulate increasingly long interaction histories, effectively retrieving and utilizing past information becomes critical for long-term interaction. Existing memory systems typically rely on semantic retrieval over compressed memories and expose predefined representations, which may miss relevant …

Review
pending
Role
unreviewed
Read
now
arxiv Score 20.0

Improved Elastic Weight Consolidation as an Optimization Constraint for Continual Learning

2026-11-11 · Davide Wiest

Research Track A · General AI

Elastic Weight Consolidation (EWC) is a widely used regularization-based technique for continual learning, protecting performance on old tasks with a quadratic penalty whose empirical Fisher is usually justified through probabilistic curvature. We instead introduce Continual Learning as an Optimization Constraint (CLOC…

Review
pending
Role
unreviewed
Read
now
arxiv Score 19.5

When Agents Need to Consolidate Memories: What RRR Consolidation Explains, and What It Does Not Explain, in Conversation ESAA

2026-11-11 · Elzo Brito dos Santos Filho

Research Track A · General AI

Language-model agents can share conversational history across sessions, but persistence does not guarantee retrieval of current decisions and their provenance. This engineering study presents Memory-TCG, a Conversation ESAA extension inspired by ringing, resonance and renormalization (RRR), and an exploratory synthetic…

Review
pending
Role
unreviewed
Read
now
arxiv Score 18.8

MemCredit: From Relevance to Utility for Decision-Time Memory in Long-Horizon Web Agents

2026-11-11 · Quanjiang Guo, Sijie Wang, Qianyang Zhu, Ben Zhang, Jinchuan Zhang, Ke Yan, Zhao Kang

Research Track B · General AI

Large language model (LLM) agents increasingly operate over long Web interaction histories, making memory management critical for reliable and efficient decision making. Existing methods primarily optimize storage, retrieval, or compression, but the memories most semantically relevant to the current state are not neces…

Review
pending
Role
unreviewed
Read
now
huggingface Score 16.0

WorldDraft: If Code Builds the World, Can Video Models Just Render It?

2026-11-12 · Zhiyu Hou, Wen-Jie Shu, Harold Haodong Chen, Weilin Ruan, Chenfei Liao, Liuzhou Zhang, Disen Lan, Xiaogang Xu, Qi Chen, Ser-Nam Lim

General AI

Multimodal coding agents decouple 4D world construction from visual generation, repurposing video models from world generators toward generative renderers. This division of labor raises a fundamental question: if code builds the world, can video models just render it? A reliable generative renderer hinges on two comple…

Review
pending
Role
unreviewed
Read
now
arxiv Score 15.8

Recover, Don't Just Continue: Learning from Feedback for Physical Failure Understanding and Correction

2026-11-11 · Miao Pan, Kaixiang Yao, Xiaohe Zhou, Xu Wang, Qinfeng Li, Wenqi Zhang, Xuhong Zhang

General AI

Multimodal large language models are regarded as the cognitive foundation of embodied intelligence, but planning from instructions alone cannot make them reliable. During execution, actions deviate from intended effects, constrained by the embodiment and environment. An embodied brain must understand why execution devi…

Review
pending
Role
unreviewed
Read
now
arxiv Score 15.8

AURA 2.0: From User-Driven Responses to Model-Initiated Conversations in Streaming Video

2026-11-12 · Shuhan Li, Huankang Guan, Xintong Guo, Yang Bo, Yuyang He, Dunyuan Xu, Jingyan Chen, Fang Liu, Tsz Kin Wan, Jinpeng Chen, Zhenqi He, Yuxuan Hu, Wenbo Li, Han Gao, Qijun Li, Long Chen, Hongsheng Li, Rui Liu

General AI

Most AI assistants respond only when prompted, remaining silent even when events in the video stream offer meaningful opportunities for conversation. To move beyond purely user-driven interaction, we present AURA 2.0, a streaming video large language model that introduces three core upgrades to AURA. The key advance is…

Review
pending
Role
unreviewed
Read
now
arxiv Score 15.8

Learning When Diffusion Language Models Should Regenerate

2026-11-12 · Minhae Oh, Jeonghye Kim, Jinwoo Koo, Yeongoon Kim, Jimin Seo, Minsik Seol, Jungwoo Lee

General AI

Remasking allows diffusion language models to revisit generated text, but effective correction requires knowing when and what to revise. Our diagnosis on Sudoku reveals two gaps. Hidden states contain error information that token confidence does not reliably expose, and remasking an isolated error usually reproduces it…

Review
pending
Role
unreviewed
Read
now
arxiv Score 15.8

XBot-Omni: An Omni-Modal Model for Embodied Understanding and Interaction

2026-11-12 · Kerui Gu, Xing Fu, Shuoyang Sun, Cheng Peng, Jingru Duan, Huayu Gu, Zhenyu Chen, Jiaxin Huang, Cheng Wen, Rui Zhen, Kairui Chen, Zhihui Peng, Shaoli Huang

General AI

Embodied interaction requires an agent to integrate multimodal perception with verbal and nonverbal behaviors to effectively communicate its intentions, including speech, body motion, and facial expressions. We present XBot-Omni, a Thinker--Talker--Actor architecture that extends the Omni paradigm to robotic modalities…

Review
pending
Role
unreviewed
Read
now
huggingface Score 15.0

DeepStress: Stress-Testing Deep Search Agents

2026-11-12 · Ismael Rousseau, Geraldine Damnati, Frederic Bechet

General AI

While search agents demonstrate impressive capabilities in multi-step question answering, their robustness to poor-quality evidence remains under-explored. Such evidence is rare in benchmark corpora, and resulting failures are hard to reproduce. We propose DeepStress, a dynamic stress-testing framework that subjects ag…

Review
pending
Role
unreviewed
Read
now
arxiv Score 14.8

We Are What We Remember: Mental-State-Driven Memory for Long-Term Character Dialogue

2026-11-12 · Zhicheng Jia, Shuchen Wu, Zhuoshi Tong, Haisen Liao, Sicheng Zhou

General AI

Shared history in human dialogue is not merely a repository of facts, but the foundation of evolving relationships and mutual understanding. However, mainstream conversational memory systems rely on passive semantic similarity over external fact stores, frequently retrieving topically similar records while missing rela…

Review
pending
Role
unreviewed
Read
now
arxiv Score 14.0

Scale-Induced Semantic Shift in Visual Prompt Injection Against VLMs and Web Agents

2026-11-12 · Yiheng Zhao, Runze Ji, Chenyu Yi, Yi Yu, Bingquan Shen, Alex Kot, Xudong Jiang

Research Track B · General AI

Visual content in multimedia systems may be resized, compressed, or re-rendered before it is interpreted by a vision-language model. We study a security failure mode in which these transformations change the machine-readable semantics of an image: a textual instruction is not recovered at a $512\times512$ presentation …

Review
pending
Role
unreviewed
Read
now
arxiv Score 13.8

AgriSense: Vision-Language Decision Support and Robotic Intervention in a Virtual Greenhouse Copy title

2026-11-12 · Lyes Saad Saoud, Oualid Doukhi, Reza Ghorbani, Moussa Ayyash

General AI

Autonomous greenhouse management increasingly requires robotic systems that integrate perception, navigation, environmental regulation, and plant-level intervention. This paper presents AgriSense, a framework combining an Isaac Sim virtual greenhouse, a quadruped mobile manipulator, RGB and LiDAR-inertial sensing, envi…

Review
pending
Role
unreviewed
Read
now
arxiv Score 13.8

Where LLM Agents Get Stuck Debugging Databases: Access More Than Reasoning

2026-11-12 · Gustavo Pinto

General AI

Large language model (LLM) coding agents are increasingly asked not only to write code but to debug running systems from live runtime data. We report a field observation of one such agent (Claude Code) as it diagnosed eight database incidents in a production message-dispatch engine. In every incident whose root cause d…

Review
pending
Role
unreviewed
Read
now
huggingface Score 13.0

Open Vision Banana: A Fully Open Recipe for Generative Visual Perception

2026-11-11 · Arpit Jadon, Valentinos Pariza, Jona Ruthardt, Ryousuke Yamada, A. Sophia Koepke, Andrei Bursuc, Yuki M. Asano

General AI

Recent work suggests that pretrained image generators encode real-world understanding beyond image synthesis. Vision Banana exploits this by framing vision tasks as conditional image editing. However, its data, weights, code, and implementation details are not public. We present Open Vision Banana (OVB), a fully open a…

Review
pending
Role
unreviewed
Read
now
arxiv Score 12.8

GuardianAgentFlow: Multi-Agent Coordination for Web Privacy Protection

2026-11-11 · Ruiyi Yang, Gayathri Lihinikaduarachchi, Rahat Masood, Flora D. Salim, Salil S. Kanhere

Research Track B · General AI

Web agents need to use private context without unnecessarily disclosing it. Effective protection must distinguish between requests that may be sent after sanitization and those that must be withheld, while preserving the information needed for authorized tasks. We present GuardianAgentFlow, a multi-agent framework that…

Review
pending
Role
unreviewed
Read
now
arxiv Score 12.8

AgentSSL: Can MLE Agents Leverage Unlabeled Data?

2026-11-12 · Ethan Lin, Akanksha Sarkar, Ziang Liu, Kristin Branson, Jennifer J. Sun

General AI

Applying machine learning systems to new domains introduces bottlenecks in both labeled data acquisition and manual pipeline engineering. While recent MLE agents show promise in automating pipeline design, they have been studied almost exclusively in fully supervised settings, leaving them untested in realistic regimes…

Review
pending
Role
unreviewed
Read
now
arxiv Score 12.8

K2V: Generative Keyframing For Physically Plausible Video Generation

2026-11-12 · Felix O'Mahony, Daniel Olmeda Reino, Roberto Cipolla, Ayush Tewari

General AI

Video diffusion models are able to generate samples with high-quality appearance and textural qualities. However, content produced by these methods is known to exhibit significant physical errors. This is because, in a classical video generation setting, all physical reasoning must take place simultaneously and exclusi…

Review
pending
Role
unreviewed
Read
now
arxiv Score 12.8

Scalable Conversational Assessment: Chat-Based Exams with a Two-Agent LLM Pipeline

2026-11-12 · Geoffrey Challen, Craig Zilles

General AI

Oral exams are compelling because (1) students can be asked to provide their reasoning for an answer and (2) their interactive nature permits follow-up questions to resolve ambiguous answers. But oral exams scale poorly. In this paper, we present scalable conversational assessment, an LLM-based system that conducts int…

Review
pending
Role
unreviewed
Read
now
huggingface Score 12.0

JustRUA: A Controlled Study of GPT-6 Robot-Use Agents

2026-11-11 · Zefan Wang, Yufan Dang, Xuanwei Shi, Mingqi Gao, Cheng Yin, SikYuen Tam, Hanyu Liu, Yasheng Wang, Yaxi Lu, Songyang Zhang, Xin Cong, Yuan Yao

General AI

GPT-6 Astra can drive a robot arm directly through a fixed harness, and scores of such robot-use agents on RoboDojo have risen quickly since. Reported systems differ in harness and test subset and often change several designs at once, so it is hard to tell which choices produce the gains. We take these designs apart an…

Review
pending
Role
unreviewed
Read
now
huggingface Score 12.0

SO-Arena: Benchmarking the Capability Spectrum and Reliability of System One Models

2026-11-11 · Lichen Li, Ziping Dong, Sicheng Feng, Qi Li, Xinchao Wang

General AI

System One models promise fast, general-purpose decision-making without lengthy token generation. However, understanding their capability spectrum and decision reliability requires more than aggregate performance on current benchmarks. Existing evaluations often provide limited insight into the breadth of model capabil…

Review
pending
Role
unreviewed
Read
now
huggingface Score 12.0

Actor, Judge, or Teacher? Understanding Decision Models in Long-Horizon Agents

2026-11-12 · Haochen Liu, Qianhao Zhu, Zixuan Wang, Dingming Li, Tongbo Chen, Peng Qiu, Yingqi Qu, Weiming Lu, Jun Xiao, Yongliang Shen

General AI

Decision models answer a question with a typed output, such as a choice among options with a probability for each, so one model can choose an agent's actions, evaluate actions already taken, or score every action to teach a policy with reinforcement learning. Prior work has studied them mainly as actors, judging their …

Review
pending
Role
unreviewed
Read
now
arxiv Score 11.8

Global Fit Inference for LISA with samsara++

2026-11-12 · Lorenzo Valbusa Dall'Armi, Gabriele Astorino, Riccardo Buscicchio, Angelo Ricciardone, Walter Del Pozzo

General AI

The Laser Interferometer Space Antenna (LISA) will observe a superposition of gravitational-wave signals, requiring joint inference of an unknown number of sources, parameters, and instrumental noise. To disentangle these contributions, that overlap in both time and frequency, a Global Fit (GF) analysis is needed. We p…

Review
pending
Role
unreviewed
Read
now
arxiv Score 11.8

HEIR: Harness Egocentric Intent for Human-Robot Interactions

2026-11-12 · Yichi Zhang, Bo Wang, Zhenghan Zhu, Yutong Han, Jiayin Zou, Gao Kejun, Chang Liu, Shaoxuan Xie, Chi Zhang, Shi Haoyang, Zongxuan Lai, Sen Cui, Guocai Yao, Miao Liu

General AI

For robots to assist, serve, and collaborate with humans, they must not only interpret human requests conveyed through language and multimodal intentional cues, but also execute long-horizon tasks that interleave navigation and manipulation. In this paper, we introduce HEIR, a resource for studying how to Harness Egoce…

Review
pending
Role
unreviewed
Read
now
arxiv Score 11.8

LiveGeo: Benchmarking Geographic Agents in Continuously Evolving Worlds

2026-11-12 · Pengcheng Zhou, Xuanyu Liu, Yanchen Yin, Tianjie Ju, Bobo Li, Shengqiong Wu, Hao Fei, Mong-Li Lee, Wynne Hsu

General AI

We introduce GeoWorld, a controllable and continuously evolving geographic environment platform for LLM agents, and LiveGeo, a benchmark for evaluating reliable decision-making in real-world geographic environments using GeoWorld. Unlike existing evaluations that largely reason and plan over fixed geographic states, Ge…

Review
pending
Role
unreviewed
Read
now
arxiv Score 11.8

Report of the 8th LSVOS Challenge: Mobile Unified Multimodal Understanding

2026-11-12 · Chang Liu, Henghui Ding, Tong Yang, Dayoung Kil, Seong-heum Kim, Qilin Xiang, Ronghao Zheng, Ziyang Zhang, Shanlin Jiang, Ruohan Shi, Qilin Fan, Chengfeng Qiu, Kaifeng Wei

General AI

This report summarizes the Mobile Unified Multimodal Understanding (MUMU) track of the 8th Large-scale Video Object Segmentation (LSVOS) Challenge, held in conjunction with ECCV 2026. MUMU asks for a single unified model that jointly performs multi-concept image tagging over quality, scene, and event vocabularies, open…

Review
pending
Role
unreviewed
Read
now
arxiv Score 11.5

SPIRE: Task-agnostic embodied continual learning through autonomous situation inference and causal motor-primitive reuse

2026-11-10 · Yiyang Jia, Mingyang Lyu, Gongce Cao, Feifei Zhao, Yinqian Sun, Yi Zeng

Research Track A

Embodied continual learning requires controllers to infer changing situations, retain prior behavior and reuse motor skills without task labels or known boundaries. The challenge is deciding when to recover earlier behavior and which skills can support new learning. We introduce Situation-guided Primitive Integration f…

Review
pending
Role
unreviewed
Read
now
huggingface Score 11.0

Evidence-Guided Recovery for AI Agents with Ambiguous Tool Outcomes and Partial Execution

2026-10-30 · Genliang Zhu, Xiaobo Chen, Ziyuan Wang, Chu Wang

General AI

Tool timeouts can leave AI agents facing ambiguous tool outcomes and partial execution: an action may have failed, committed or remained in flight while other workflow steps have already completed. We formulate evidence-guided task recovery over approved obligations, provider capabilities and states compatible with pub…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 10.8

Broadcast, Don't Replicate: Eliminating Weight HBM in Datacenter LLM Serving

2026-11-12 · Hannah Atmer, Yuan Yao, Thiemo Voigt, Stefanos Kaxiras

General AI

Large language model (LLM) inference at datacenter scale is fundamentally bottlenecked by memory capacity and bandwidth limits. Because model weights are read-only and identical across instances, conventional architectures that replicate these parameters across the local DRAM of tens of thousands of accelerators incur …

Review
pending
Role
unreviewed
Read
now
arxiv Score 10.8

Classical Energies Learn to Reason 1,000x Faster than Looped Models

2026-11-12 · Benjamin Hoover, Hugh Blayney, James C. R. Whittington

General AI

Looped models have emerged as a prominent paradigm for reasoning, with successes on simple reasoning puzzles such as Sudoku-Extreme and Maze-Unique. However, these models can be unstable, slow to train, and use many parameters and FLOPs to solve simple puzzles. Descent on an energy not only mitigates these issues but a…

Review
pending
Role
unreviewed
Read
now
arxiv Score 10.8

CoReMem: Cost-Internalizing Memory for Successive Coding Tasks with Prospective Valuation

2026-11-12 · Peiwen Li, Harshit Verma, Sophia Zhou, Haiwen Wang, Ngoc Bui, Heng Chang, Fei Fei, David van Dijk, Rex Ying

General AI

Coding agents tackle successive tasks on an evolving repository, carrying forward edits and writing memory before the next task is known. Selecting memory for unknown next tasks requires balancing future usefulness against the costs of writing and reading it. The actor must also decide when acting on available memory s…

Review
pending
Role
unreviewed
Read
now
arxiv Score 10.8

Metal Sandbox: Isolated Execution and Command Scheduling for GPU Coding Agents on Apple Silicon

2026-11-12 · Haozhe Wang, Lina Chen, Fanqi Wan, Qi Han, Yinmin Zhang, Zheng Ge, Daxin Jiang, Junchi Yan

General AI

On-device inference requires kernels that make effective use of local accelerators. Efficient training and evaluation of models that generate Apple Metal kernels benefit from native macOS tools and concurrent agent sessions. These sessions alternate between model requests, development commands, and GPU execution, leavi…

Review
pending
Role
unreviewed
Read
now
arxiv Score 10.8

Unbundling Multimodal Vision Encoding

2026-11-12 · Ellis Brown, Chris Hoang, Shengyi Qian, Shang-Wen Li, Mike Lewis, Marjan Ghazvininejad, Luke Zettlemoyer, Lucas Beyer, Mengye Ren, Jakob Verbeek, Saining Xie, Hu Xu

General AI

How multimodal language models should encode visual input remains up for debate. Most models use a pretrained vision encoder, while some recent recipes train the encoder from scratch with the language model or remove it entirely. Vision encoders bundle together three design choices: a visual prior from upstream pretrai…

Review
pending
Role
unreviewed
Read
now
arxiv Score 10.5

Janus-MOPD: Tracing the Student Learning Dynamics in Multi-Teacher On-Policy Distillation

2026-11-11 · Zongji Yu, Ming Zhang, Dongcheng Zhang, Jiawei Zhou, Haotian Pang, Yiliu Sun, Xiang Wang, Youbang Sun, Chaochao Lu, Chen Gong

Research Track A · General AI

Multi-domain capabilities are critical for practically applicable Large Reasoning Models (LRMs). Therefore, Multi-teacher On-Policy Distillation (MOPD) has become a crucial part of the post-training process, integrating the capabilities of multiple domain-specific teacher models into a single student model. However, th…

Review
pending
Role
unreviewed
Read
soon
huggingface Score 10.0

Factored One-sided Adam-Moment for Practical and Scalable SOAP

2026-11-12 · Jaemyung Yu, Byeongho Heo, Sangdoo Yun, Dongyoon Han

General AI

Second-order optimizers (e.g., Shampoo, SOAP) have emerged as compelling alternatives to Adam/AdamW and promise faster per-step convergence than AdamW in transformer training. However, they still remain costly in practice, and even SOAP still pays substantial optimizer-state memory and eigendecomposition overhead. We a…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 10.0

KuaiRecV2: Benchmarking Large-Scale Continual Learning for Diversified and Multi-task Recommendation

2026-11-12 · Chenxu Li, Shuchang Liu, Hantao Shu, Wei Yuan, Zhe Xu, Guoqing Hu, Yan Wang, Biao Yang, Xiang Li, Cheng Ling, Fan Yang, Yongqi Liu, Lantao Hu, Han Li, Kaiqiao Zhan, Kun Gai, An Zhang, Xiang Wang

Research Track A · General AI

Recommender systems, which retrieve items of interest to users based on their interaction histories, have recently witnessed several key technological advancements, benefiting from the semantic representation space and model-data scaling. Nevertheless, practitioners and researchers have frequently encountered critical …

Review
pending
Role
unreviewed
Read
soon
arxiv Score 9.8

Al-Ghasali: A Computational Framework for Simulating Islamic Jurisprudential Rulings via Finite State Automata and Large Language Models

2026-11-12 · Elnaserledinellah Mahmoud Abdelwahab

General AI

This paper presents Al Ghasali, a computational framework designed to formalize and simulate Islamic jurisprudential (Fiqh) rulings. Building upon recent set theoretic formalizations of al sabr wa al Taqsim that extract minimal operational rules as simple Horn clause logical programs, this work proves that the semantic…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 9.8

ArchSpace: Recording, Reviewing, and Referencing LLM Architecture Research

2026-11-12 · Jie Wang, Tong Zhu, JiaHao Kuang, Zhe Xu, Yunkai Chai, Dehong Zheng, Yu Cheng, Tao Ji, ZhouHan Lin, Qi Zhang, Bowen Zhou

General AI

Model architecture shapes the scaling behavior, training efficiency, and inference cost of LLMs, making it a central focus of both industry and academia. In industry, teams such as those behind DeepSeek, Qwen, and Kimi routinely introduce architectural innovations, yet public reports rarely document which ideas succeed…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 9.8

Internalizing Investment Mandates: A Neurosymbolic RL Approach to Portfolio Allocation

2026-11-12 · Zeynep Yıldız, Satchit Chatterji, Ercan Atam, Erman Acar

General AI

Institutional portfolios operate under investment policy statements that specify asset allocation constraints, often conditioned on market regime information. Deep reinforcement learning has been shown to optimize financial performance but typically does not enforce rule compliance, making it unsuitable for real-world …

Review
pending
Role
unreviewed
Read
soon
huggingface Score 9.4

Learning the Task or Learning the Verifier? Rethinking Criteria Exposure in Embodied Self-Improvement

2026-11-05 · Wenbo Huang, Jianxin Bi, Shunyu Yang, Ruiyang Si, Rui Ni, Qiang Wang, Shulong Jiang, Yintianrun Zhang, Feixiang Ruan, Kewei Zhang, Ye Huang, Shukai Gong, Daquan Zhou

General AI

Self-improving embodied agents are increasingly being developed and compared on robot simulation benchmarks. Several recent works expose the benchmark's success criteria to the learning agent, a practice we call Criteria Exposure (CE). These criteria, originally written to verify policies trained by imitation learning,…

Review
pending
Role
unreviewed
Read
soon
huggingface Score 9.0

UniSpark: Exploring Hybrid Attention for Multimodal Understanding and Generation

2026-11-12 · Jiancheng Pan, Yuzheng Liu, Hengyuan Cao, Qiyuan Zhang, Yangyan Li, Biao Gong, Zheng Zhang, Ka Long Cheng, Xinyang Wang, Yujun Shen

General AI

We present UniSpark-9B, a unified multimodal model that investigates hybrid attention for visual understanding and generation. Built on a Mixture-of-Transformer-Experts backbone with approximately 9B parameters in each branch, UniSpark combines full attention for explicit token interactions with GatedDeltaNet linear at…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 8.8

Interactions from memory footprints: Cage-hopping, mnemotaxis & nonreciprocity

2026-11-12 · Jyoti Prasad Banerjee, Sambodh Sinha, Shashi Thutupalli, Rituparno Mandal, Deb Sankar Banerjee

General AI

Cellular motility is integral to biological processes ranging from foraging to immune responses and cancer metastasis. While cells often navigate by sensing pre-existing environmental cues, many actively steer their future trajectories by remodeling their surroundings via shared, long-lived physicochemical trails calle…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 8.8

Sketching Language Models for Context Attribution

2026-11-12 · Christopher Liaw, Aranyak Mehta, Grigoris Velegkas, Felix Zhou

General AI

Context attribution asks how removing parts of a prompt changes a language model's output. We study leave-one-out (LOO) attribution for a prompt of $n$ tokens and a fixed measurement function. Computing all LOO values directly requires $n$ model evaluations. Under a stability assumption, our algorithm estimates all LOO…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 8.8

Sustainable Federated Fine-Tuning with LoRA: Environmental Impact and Performance Footprint

2026-11-12 · Pedro Augusto Londe Ribeiro, Rodrigo Moreira, Larissa Ferreira Rodrigues Moreira, Joberto S. B. Martins, Tereza C. Carvalho, Flávio de Oliveira Silva

General AI

Federated Learning (FL) with Parameter-Efficient Fine-Tuning (PEFT) enables efficient adaptation of Large Language Models (LLMs), yet its energy and carbon costs across hardware remain poorly understood. We evaluated federated Low-Rank Adaptation (LoRA) fine-tuning on NVIDIA A40, RTX 6000 Ada, and A100 GPUs using a $2^…

Review
pending
Role
unreviewed
Read
soon
huggingface Score 8.0

Do Multimodal Models Decide Consistently on Equivalent Optical Spectra?

2026-10-30 · Ziyuan Wang, Genliang Zhu, Xiaobo Chen

General AI

A spectrum can be given to a model as a table in nanometers or as a plot in photon energy, and the decision should not depend on which one it sees. When the data needed for an answer are missing, the model should say so. We test representation invariance and abstention on measured optical spectra of silicon metasurface…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 7.8

Context-Predictive Low-Rank Adaptation for Face Presentation Attack Detection

2026-11-12 · Aminu Yusuf, Shakeel Muhammad Ibrahim, Sulaimon Oyeniyi Adebayo, Muhammad Sohail Ibrahim, Muhammad Shahid Jabbar, Shujaat Khan

General AI

Face presentation attack detection requires representations that distinguish physical spoofing cues across changes in acquisition devices, subjects, and attack media. Large visual encoders offer transferable representations, but conventional binary fine-tuning supervises their final decisions without explicitly constra…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 7.8

Fictitious Play When Agents Are Not Expected-Utility Maximizers: A Geometric Analysis in $2$-Action Games

2026-11-12 · Hao Liang, Solaleh Mohammadi, Kaiqing Zhang

General AI

Fictitious play (FP) is one of the earliest and most fundamental learning procedures in games, where each agent independently chooses a best response to the empirical distribution of opponents' past actions by maximizing the expected payoff with respect to it. However, learning agents may not always be expected-utility…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 7.8

PACER: GPU Excursion Management for ML Workloads

2026-11-12 · Rutwik Jain, Nathaniel Morris, Yiwei Jiang, Shidhartha Das, Gary Kao, Indrani Paul, Shivaram Venkataraman, Srilatha Manne, Matthew D. Sinclair

General AI

Variability is an inherent characteristic of modern GPU systems, arising from a combination of process variation, dynamic environmental conditions, and power management decisions. Prior work has characterized inter-GPU variability and found consistent performance differences across GPUs. However, we observe that modern…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 7.8

Preserve the Vulnerability, Break the Alert: Testing the Robustness of CodeQL Taint Tracking with Natural Rewrites

2026-11-12 · Zhezhen Cao, Weiyuan Ding, Bowen Xu

General AI

Static application security testing (SAST) tools such as CodeQL serve as security gates in continuous integration and code review, and a gate is also read in reverse. When a previously reported alert disappears after a change, developers and the tools around them often take the vulnerability to be gone. That inference …

Review
pending
Role
unreviewed
Read
soon
arxiv Score 7.3

Programming Quantum Circuit Parameters with Large Language Models

2026-11-12 · Tiago de Souza Farias, Caio Moraes Porto, Matheus da Silva Fonseca, Nicolas Armando Cabrera Carpi, Anderson de Araujo Oliveira, Celso Jorge Villas-Boas

General AI

Large language models (LLMs) have recently demonstrated the ability to generate parameters for variational quantum circuits without conventional optimization. However, the conditions under which this approach is effective remain unclear. Here, we investigate a frozen, general-purpose LLM that generates complete circuit…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 6.8

Assessing the Reliability of Longitudinal Bibliometric Rankings under Incomplete Citation Histories

2026-11-12 · João Paulo Papa

General AI

Longitudinal bibliometric analyses rely on annual citation histories, yet missing annual observations are frequently difficult to distinguish from true zero-citation years. We present a reproducible case-study framework for auditing citation-history gaps, tracing retrieval provenance, and quantifying the sensitivity of…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 6.8

Ballad: Bandit-Based LLM Routing for Automated Heuristic Discovery

2026-11-12 · Samidha Verma, Ankit Anand, Sayan Ranu

General AI

Modern automated heuristic discovery (AHD) systems increasingly rival hand-crafted heuristics by leveraging Large Language Models (LLMs) to iteratively mutate and refine solutions. A natural assumption is that more powerful and more expensive models consistently yield better heuristics. We challenge this empirically: a…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 6.8

Spatial Foundry: A Multimodal Dataset for Spatial Intelligence

2026-11-12 · Ethan Lin, Yang Song

General AI

Spatial intelligence requires connecting visual observations to the geometry, objects, and persistent state of an environment. Images alone leave these relationships implicit, while disconnected annotations make it difficult to study appearance, camera motion, and scene changes under a common reference. We introduce Sp…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 5.8

Back to Pixels: Lifting Latent Video Diffusion into Pixel Space

2026-11-12 · Yanke Song, Shuo Chen, Chris Rockwell, Guande He, Nan Liu, Desai Xie, Tianpei Gu, Yong-Hyun Park, Aryan V S, Dennis Yuren Menn, Ming-Hsuan Yang, Xun Huang

General AI

Frontier video generation models work in a VAE latent space that compresses video in both space and time. This compression made training tractable, but now limits high-resolution, controllable, and interactive generation -- doubling resolution quadruples token count, conditions apply to groups of frames rather than ind…

Review
pending
Role
unreviewed
Read
later
arxiv Score 5.8

Detection Is Not Early Warning: Evaluating Token-Level Alarms on Language Model Failures

2026-11-12 · Marco De Negri, Luca Sartori, Lorenzo Baggi, Moritz Reihs, Eduard Frank Ďurech, Mennatallah El-Assady, Valentina Pyatkin, Anna Hedström

General AI

Many failures of language models begin at a specific token, the onset. A monitor that reads the model during generation can classify the finished response, detect the failure once it is under way, or warn before it begins, and only a warning leaves time to stop or steer the generation. Monitors are usually evaluated wi…

Review
pending
Role
unreviewed
Read
later
arxiv Score 5.8

TTB: Test-time Baking for Efficient Rendering of Decoder-only View Synthesis Models

2026-11-12 · Chin-Yang Lin, Ryo Hachiuma, Jaesung Choe, Min-Hung Chen, Yu-Chiang Frank Wang, Yen-Yu Lin, Wei-Chen Chiu, Yu-Lun Liu, Cheng Sun

General AI

Decoder-only Large View Synthesis Models (LVSMs) with KV-cache have recently achieved state-of-the-art quality by regressing novel views from neural networks without reconstructing 3D geometry, while Test-Time Training (TTT) layers save computation but sacrifice quality. We propose Test-Time Baking (TTB), a strategy th…

Review
pending
Role
unreviewed
Read
later