Research Paper Cockpit

Daily Digest - 2026-08-10

Papers first seen in this daily snapshot.

Daily Archives

Quick jump into generated daily digests.

Research Workflow

Latest digest: 2026-08-18.

Papers

50 visible entries

arxiv Score 23.2

ResidencyRL: Reinforcement Learning in Simulated Clinical Environments

2026-08-07 · Valentin Liévin, Samuel Schmidgall, Tim Strother, Alex Bijamov, Akshay Goel, Anil Palepu, Chunjong Park, Vahid Balazadeh, Min Woo Sun, Marius Guerard, Justin Chen, Dave Steiner, Vikram Dhillon, Ibrahim Azar, Akhil Mehta, Nicholas Spetsieris, Shilpan Shah, Maen Abdelrahim, Amit Dahiya, Yun Liu, Katherine Chou, Yossi Matias, Avinatan Hassidim, Dale R. Webster, Quoc V. Le, Raia Hadsell, Joelle Barral, Carey Radebaugh, Aleksandra Faust, Shekoofeh Azizi, Mike Schaekermann, Po-Hsuan Cameron Chen, Tao Tu, David Racz, Lin Yang

General AI

In medical education, physicians convert academic knowledge into clinical expertise through residency: years of training across thousands of encounters, with diverse sources of feedback and progressively greater autonomy. Much of clinical reasoning relies on the patient encounter, a dialogue in which a clinician elicit…

Review
pending
Role
unreviewed
Read
now
arxiv Score 20.2

Fisher-R1: Training LLM Agents for Reliable Hypothesis Testing

2026-08-07 · Jiacheng Miao, Jin Mu, Guanhua Chen, James Zou

General AI

Reliable hypothesis testing is the foundation of many empirical scientific claims. Large language model (LLM) agents are increasingly used to automate this process, as they can inspect datasets, generate code, and produce analyses end-to-end. However, we show that they frequently make subtle inferential errors that lea…

Review
pending
Role
unreviewed
Read
now
huggingface Score 19.0

Douyin Multimodal Embedding Model Technical Report

2026-08-03 · Haonan Chen, Chu Li, Zhicheng Wang, Yuanwei Liu, Yuanjiang Wang, Shaohua Jiang, Zhicheng Dou

General AI

Multimodal representation learning is a cornerstone of modern AI. By encoding multimodal queries and targets into vectors, it powers industrial search and recommendation and underpins modern agents. Real-world platforms with complex modalities and massive-scale content, such as Douyin, Xiaohongshu, and YouTube, demand …

Review
pending
Role
unreviewed
Read
now
huggingface Score 18.4

StreamArena: Toward Continuous, Interactive, and Long-Horizon Agentic Streaming Video Understanding

2026-08-06 · Xichen Zhang, Guankai Li, Yinghao Zhu, Shijian Wang, Sitong Wu, Shaozuo Yu, Meng Chu, Yuan Lu, Jiaya Jia

General AI

Deploying autonomous multimodal agents in continuous, real-world environments requires them to ingest unbounded audio-visual streams and maintain hour-scale memory. However, current evaluations predominantly rely on brief clips and multiple-choice formats. This design allows minimal baselines that process only the last…

Review
pending
Role
unreviewed
Read
now
arxiv Score 18.2

An AI4AI Framework for Visual Token Pruning

2026-08-07 · Zhen Liu, Wenli Huang, Wei Song, Yuhan Liu, Zhiqin Yang, Jingwen Fu

General AI

Visual-token pruning can substantially reduce the inference cost of multimodal large language models (MLLMs), yet existing methods largely rely on fixed, handcrafted heuristics and costly expert trial and error. As pruning objectives, budgets, and model architectures diversify, manually navigating the expanding design …

Review
pending
Role
unreviewed
Read
now
arxiv Score 18.2

Toward a Causal Data Management Ecosystem for Decision Making and Agentic AI

2026-08-07 · Dazhuo Qiu, Yingli Zhou, Amedeo Pachera, Angela Bonifati, Andrea Mauri

General AI

Modern AI is no longer a single model but an ecosystem: classical ML predictors, deep and multimodal models, large language models, and agents, each trained and tuned over different data sources and each producing outputs at scale that become inputs to the others. Operating such an ecosystem is fundamentally a data int…

Review
pending
Role
unreviewed
Read
now
arxiv Score 17.2

GeoBenchLLM: A Comprehensive Benchmark for Evaluating LLMs on Geo-Related Tasks

2026-08-07 · Rodrigo Ferreira Rodrigues, Karim Radouane, Jose G Moreno, Lynda Tamine

General AI

In the context of geodata, existing Large Language Models have often been studied in a homogeneous setting, which has considerably limited insights into their generalization capabilities. In this paper, we present \benchName, a comprehensive benchmark for probing LLMs on geo-related tasks. We leverage a careful selecti…

Review
pending
Role
unreviewed
Read
now
arxiv Score 16.2

I Seek You in Videos: Identity-Conditioned Queries for Person-Centric Video Reasoning

2026-08-07 · Shibo Gao, Chongxiao Wang, Chenglong Huang, Jie Ma, Haolin Shi, Fei Ding, Jing Li, Qiang Lyu, Yangyang Liu, Yang Liu, Jun Liu, Linlin Huang, Peipei Yang

General AI

Real-world video reasoning often involves multimodal, multi-source inputs, whereas existing video reasoning tasks typically assume a simplified video-text setting, limiting identity matching and person-centric reasoning. To bridge this gap, we introduce the Identity-conditioned Queries (ICQ) task, in which models are r…

Review
pending
Role
unreviewed
Read
now
arxiv Score 16.2

Strategy-first synthesis planning for complex natural products

2026-08-07 · Daniel Armstrong, Xuan-Vu Nguyen, Octavian Susanu, Gabriel Gibberd, Théo A. Neukomm, Taddäus Strunden, Dan Forster, Morgane Delattre, Shawn Teh, Clément Rols, John Federice, Hayden Leatherwood, M. Lavelle Barnes, Maarten R. Dobbelaere, Peter Wipf, Jon T. Njardarson, Jieping Zhu, Philippe Schwaller

General AI

The total synthesis of a complex molecule is among the most demanding intellectual and experimental feats in chemistry: a chemist must plan many steps ahead for how to assemble simple building blocks into an intricate target, devise backup strategies, and anticipate procedural challenges. It is also a profoundly creati…

Review
pending
Role
unreviewed
Read
now
huggingface Score 15.4

The Optimizer Is the Agent: Reasoning-Driven Search across Prompts, Programs, and ML Workflows

2026-08-07 · Junbo Li, Boyi Liu, Canwen Xu, Yite Wang, Yuxiong He, Zhangyang Wang, Qiang Liu, Zhewei Yao

General AI

Recent systems for optimizing prompts, programs, and ML workflows typically rely on explicit outer-loop controllers such as evolutionary search, bandits, or textual-gradient methods. We ask a fundamentally different question: how much of this search policy can be internalized by a single tool-using agent? We present Re…

Review
pending
Role
unreviewed
Read
now
arxiv Score 15.2

PsychoAgent: An Affect-Sensitive Cognitive Architecture for Conflict-Aware Memory in LLM Agents

2026-08-07 · Mohammad Amanlou, Parham Abed Azad, Farbod Davoodi, Mostafa Masumi, Behnam Bahrak, Abdol-Hossein Vahabie

General AI

Human-like cognition does not select past experience by topical similarity alone: affective significance and unresolved conflict also shape what becomes accessible. We present PsychoAgent, a cognitive architecture for LLM agents that separates factual and affective memory and integrates both through a conflict-aware ex…

Review
pending
Role
unreviewed
Read
now
arxiv Score 14.9

WebRider: Persona-Conditioned Intent Controllers for Live-Web Assistance

2026-08-07 · Zhi Li, Tao Zhou, Yeqing Li, Eugene Ie, Demetri Terzopoulos

Research Track B · General AI

Delegating a web task involves more than asking a question; it requires transferring a policy: what to verify, how to handle uncertainty, which preferences matter, and when to stop. Yet, current live-web agents are evaluated solely on the final answer, ignoring the policy constraints that define the delegation. A plaus…

Review
pending
Role
unreviewed
Read
now
arxiv Score 14.2

TEPA: Revoking Stale Memories for Conflict-Robust Language Agents

2026-08-07 · Yan Zhou, Yue Ouyang, Kaiyang Zheng, Suncheng Xiang

Research Track A · General AI

Long-term memory enables language agents to reuse past facts, preferences, and task experience. Persistence also creates a central falsifiability problem: when the world changes, stale memories can remain retrievable and pollute the prompt. We characterize this failure mode as memory pollution: degradation caused by ac…

Review
pending
Role
unreviewed
Read
now
huggingface Score 13.4

SFT Conflicts, RL Coexists: A Theoretical and Empirical Analysis of Multi-Task Learning for LLMs

2026-08-06 · Kejian Zhu, Zhuoran Jin, Shangqing Tu, Hongbang Yuan, Yushi Bai, Kang Liu, Juanzi Li, Jun Zhao

General AI

Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL) exhibit fundamentally different behaviors in enhancing multi-task reasoning for large language models (LLMs). Our preliminary experiments revealed a phenomenon: SFT suffers from severe task conflicts under multi-stage training, whereas RL enables stable coexi…

Review
pending
Role
unreviewed
Read
now
arxiv Score 13.2

Blast Radius

2026-08-07 · MY Pitsane, Hope Mogale

Research Track A · General AI

Agentic coding faces growing problems of affordability and wasted tokens. We introduce Blast Radius, a predictive memory management layer that estimates an incoming prompt's reach through coupled context and code channels. NECROPHORESIS enables reversible eviction by archiving dead context verbatim, while Recurring Dea…

Review
pending
Role
unreviewed
Read
now
arxiv Score 13.2

WNM-3D: A World Navigation Model with 3D Scene Conditioning for Closed-Loop VLN

2026-08-07 · Yuehao Huang, Yunzi Wu, Xiaotao Zhang, Xinhai Li, Jiankun Dong, Jiajun Lv, Chi Zhang, Chenjia Bai, Yong Liu, Xuelong Li

General AI

Recent vision-language navigation (VLN) systems increasingly adapt pretrained vision-language models (VLMs) into vision-language-action (VLA) policies that map egocentric observations and language instructions directly to navigation actions. Although semantically capable, such action-centric training does not explicitl…

Review
pending
Role
unreviewed
Read
now
arxiv Score 12.2

CoBa: Cost-Effective Test-Time Scaling via Compute-Balanced Routing

2026-08-07 · Yan Zhou, Yue Ouyang, Kaiyang Zheng, Suncheng Xiang

General AI

Test-time scaling is often implemented by spending more compute along one axis: sampling more solutions, extending a chain of thought, or applying a stronger evaluator. Under a fixed inference budget, these choices compete. This paper formulates test-time reasoning as a compute-allocation problem in which a system must…

Review
pending
Role
unreviewed
Read
now
arxiv Score 12.2

CreativeInstruct: Scalably Teaching LLMs to Balance Quality, Creativity, and Diversity

2026-08-07 · Ananya Sahu, Mohit Bansal, Elias Stengel-Eskin

General AI

While post-training improves the capabilities of large language models (LLMs), it generally lowers their output diversity and creativity, negatively impacting tasks that explicitly require creativity (e.g., story generation) as well as those that require it implicitly, e.g., reinforcement learning (RL). We instead prop…

Review
pending
Role
unreviewed
Read
now
arxiv Score 12.2

LoRAScan: Detecting Backdoor Prompts in Low-Rank Adapters for Large Language Models via Down-Projection Activation Spikes

2026-08-07 · Doniyorkhon Obidov, Honggang Yu, Xiaolong Guo, Kaichen Yang

General AI

Low-rank adaptation (LoRA) enables efficient specialization and distribution of large language models through compact adapters. However, untrusted adapters introduce a supply-chain threat: a backdoored adapter can cause a model to generate harmful content, malicious code, political propaganda, or covert advertisements …

Review
pending
Role
unreviewed
Read
now
arxiv Score 12.2

MemPrism: Task-Conditioned Relational Memory Views for Long-Horizon Agents

2026-08-07 · Zhisheng Chen, Bingfan Zeng, Bangde Cao, Zhengwei Xie, Yuxuan Li, Jinhan Li, Zheng Lu, Xiangchen Guan, Zikai Xiao, Rui Qian, Jingwei Song

General AI

Long-horizon agents rely on memory to reuse experiences, yet existing memory systems often assume that evidence can be directly consumed through a fixed representation. This leads to representation mismatch, where relevant information is available but not organized for the current decision. To this end, we propose MemP…

Review
pending
Role
unreviewed
Read
now
arxiv Score 12.2

Token Communication for Multimodal Large Language Model

2026-08-07 · Jingkai Ying, Zhijin Qin, Yuan Shen, Khaled B. Letaief

General AI

With the broad success of the Transformer architecture, token is becoming a new basic information processing unit. This trend is especially evident in multimodal large language models (MLLMs), where both visual and textual information are represented and processed as tokens. With the rapid deployment of MLLMs, the effi…

Review
pending
Role
unreviewed
Read
now
huggingface Score 11.4

PrivacyPeek: Auditing What LLM-Based Agents Acquire, Not Just What They Say

2026-08-06 · Mingxuan Zhang, Jiahui Han, Dadi Guo, Songze Li, Guanchu Wang, Na Zou, Dongrui Liu, Xia Hu

General AI

LLM-based agents are rapidly advancing, autonomously invoking external tools to complete multi-step tasks for users. However, agents often acquire more sensitive information than the task requires. Existing privacy benchmarks audit what the agent's response or outgoing actions disclose, but overlook the acquisition sta…

Review
pending
Role
unreviewed
Read
now
arxiv Score 11.4

SkillProx: Self-Evolving Agent Skills via Proximal Textual Gradient Descent

2026-08-07 · Mingxuan Zheng, Yujin Zhou, Chuxue Cao, Boqin Yin, Yuyao Zhang, Jiapeng Sun, Shuaishuai Gong, Sirui Han, Yike Guo

Research Track A · General AI

LLM agents increasingly adapt to recurring tasks by accumulating procedural knowledge in skills. These skills are lightweight, reusable textual artifacts that are loaded into the agent's context without weight updates. Recent methods refine skills through iterative task execution, failure diagnosis, and trajectory-guid…

Review
pending
Role
unreviewed
Read
now
arxiv Score 11.2

RoRA: Role-Oriented Regional Allocation for Visual Token Pruning in MLLMs

2026-08-07 · Qiyanhui Lu, Han Wu, Rongjian Xu, Tingzhang Luo, Cheng Fan, Xinghao Chen, Minjing Dong, Jufeng Yang, Jianyuan Guo

General AI

Multimodal large language models (MLLMs) encode images as long visual token sequences, making prefilling and KV-cache storage expensive. Existing training-free pruning methods select tokens by importance, diversity, or spatial coverage, but treat retained tokens as interchangeable and do not explicitly track which obje…

Review
pending
Role
unreviewed
Read
now
arxiv Score 11.2

Taxonomy-Driven Analysis of Open-Source AI Risk Mitigation Tools

2026-08-07 · Afreen Alam, Evgenija Popchanovska, Ana Gjorgjevikj, Maryan Rizinski, Lubomir T. Chitkushev, Irena Vodenska, Dimitar Trajanov

General AI

Rapid adoption of large language models (LLMs) in enterprise settings has introduced operational, security, and governance risks. As generative AI applications move from pilot to production, manual harm identification and mitigation are becoming difficult to scale. Although many tools support model evaluation, adversar…

Review
pending
Role
unreviewed
Read
now
huggingface Score 10.4

Beyond Simply Environment Scaling: Designing Effective Environment Distributions for Multimodal Agent Learning

2026-08-06 · Kejian Zhu, Zhuoran Jin, Dongqi Huang, Hongbang Yuan, Yupu Hao, Kang Liu, Jun Zhao

General AI

Recent works train agents by constructing large-scale multimodal environment pools. However, we find that simply increasing the number of multimodal environments does not always benefit. We further analyze the limitations in current multimodal environment distributions through a series of experiments. Based on these fi…

Review
pending
Role
unreviewed
Read
soon
huggingface Score 10.4

YOLO-PEFT: Parameter-Efficient Fine-Tuning on YOLO Family

2026-08-07 · Xu Lin, WenJie Nie, Jinlong Peng, Weifu Fu, YueXiao Ma, Xiawu Zheng, Yong Liu

General AI

Generic parameter-efficient fine-tuning (PEFT) methods transferred from language models can fail silently on real-time detectors, whose heterogeneous operators and detection-specific components impose placement constraints absent from regular Transformer stacks. We propose YOLO-PEFT, a structure-aware framework that fo…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 10.2

Circuit-Based Program Verification: Sequential Circuits as an Intermediate Representation for Verifying C Programs

2026-08-07 · Po-Chun Chien, Nian-Ze Lee, Armin Biere, Dirk Beyer

General AI

Formal verification of software programs and hardware designs shares the common goal of reasoning about state-transition systems, yet the two communities have largely developed separate intermediate representations and verification algorithms. This paper investigates sequential circuits as an intermediate representatio…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 10.2

Diffusion LLMs as Targets and Adversaries: Mechanistic Safety Exploits

2026-08-07 · Elena Dumitrescu, Gert Lek, Lydia Y. Chen, Jérémie Decouchant

General AI

Diffusion Large Language Models (DLLMs) replace autoregressive next-token prediction with iterative parallel denoising, yet their internal safety mechanisms remain poorly understood. In this work, we investigate DLLMs both as targets and as adversaries, exposing mechanistic vulnerabilities in diffusion-based alignment.…

Review
pending
Role
unreviewed
Read
soon
huggingface Score 9.4

Do AI Personas Grow? Analyzing and Benchmarking Personality Evolution in LLM Agents After Life Events

2026-08-06 · Ming Wang, Peidong Wang, Xiaocui Yang, Daling Wang, Shi Feng, Fiona Fui-Hoon Nah, Ee-Peng Lim

General AI

Personality-conditioned LLM agents (PC-Agents) are increasingly used in emotional support, social simulation, and role-playing, motivating the development of lifelong agents that remain coherent over extended interactions. A key component of such coherence is personality evolution: agents should undergo plausible, psyc…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 9.2

CoinRAG: Contextualized Information Nugget KV Cache Reuse for Long-Context RAG

2026-08-07 · Gyuwan Kim, Cheoneum Park, Tao Yang

General AI

Recent optimization studies on Retrieval-Augmented Generation (RAG) have exploited chunk-level KV cache reuse to avoid processing long retrieved contexts for higher efficiency, while significant information redundancy and noise still remain in the coarse-grained chunks. This paper optimizes the Pareto frontier under lo…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 9.2

Conformal Fusion Under Missing Modalities

2026-08-07 · Alireza Moayedikia

General AI

Multimodal fusion architectures typically assume all modalities are available at inference, yet sensor failures, acquisition variability, and cost constraints routinely produce incomplete observations. Existing work treats modality absence as a prediction-accuracy problem, leaving a more basic question unanswered: whet…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 9.2

EMAS: Stabilizing Multi-Agent System Evolution through Evidence-Guided Revision

2026-08-07 · Chao Fei, Qingyi Si, Kaihua Liang, Yanghua Xiao, Panos Kalnis, Hongcheng Guo

General AI

Many methods for automated multi-agent system design optimize prompts and topologies during an initial design stage and then deploy the resulting system unchanged on subsequent samples. Experience from these samples is rarely consolidated into reusable system updates, while accuracy-oriented designs may incur high toke…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 9.2

MirrorWorld: Taming Video Diffusion Models for Mirror Reflection Generation

2026-08-07 · Youjun Zhao, Alex Warren, Gary K. L. Tam, Rynson W. H. Lau

General AI

Recent advances in video diffusion models (VDMs) have enabled high-fidelity video synthesis. However, generating mirror reflections remains challenging because the content within a mirror must remain consistent with the surrounding scene. Existing VDMs are not specifically designed to model scene-to-mirror relationship…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 9.2

SkillAligner: Treating Retrieved Skills as Adaptable Drafts at Execution Time

2026-08-07 · Qinfeng Li, Dalin He, Yuntai Bao, Ying Yang, Ruoxi Chen, Xinyan Yu, Lizhou Liang, Ge Su, Wenqi Zhang, Xuhong Zhang

General AI

General-purpose skills promise reusable procedural knowledge for language agents, yet semantic relevance does not guarantee execution utility: a retrieved skill may encode assumptions that conflict with the current task, execution environment, or other retrieved skills. We formalize this problem as the skill--execution…

Review
pending
Role
unreviewed
Read
soon
huggingface Score 8.4

Can MLLMs Decode the Creative Leap? Introducing C4 for Cross-Concept Understanding

2026-08-06 · Ming Wang, Yuqing Zhang, Tingna Xie, Xiangju Li, Xiaocui Yang, Daling Wang, Shi Feng, Yifei Zhang

General AI

Creative capabilities of MLLMs matter in design, communication, education, and human--AI collaboration, yet remain difficult to evaluate because explicit targets and reward signals are scarce compared with accuracy-oriented tasks. Cross-concept understanding is a core cognitive capacity underlying receptive creativity.…

Review
pending
Role
unreviewed
Read
soon
huggingface Score 8.0

Round-Trip Consistency: Bidirectional Diffusion Models Can Predict Their Own Rollout Errors

2026-08-01 · Alexander Scheinker

General AI

Autoregressive models accumulate error over long rollouts, yet at deployment there is no ground truth to measure it against. We train a single conditional latent diffusion model that steps a dynamical system forward or backward in time via a direction flag, and show that this bidirectionality supplies a measurement-fre…

Review
pending
Role
unreviewed
Read
soon
huggingface Score 7.4

Zero Gap Is Not Restoration: Stratified Per-Question Probability Evaluation and Step-wise Mitigation of Benchmark Contamination

2026-08-07 · Ruijie Hou, Yueyang Jiao, Zhao Wang, Yingming Li

General AI

Test data from public benchmarks inevitably leaks into pretraining corpora, inflating evaluation scores once memorized. Contamination mitigation evaluation intervenes in the decoding process to suppress memorization and restore a contaminated model's genuine capability, but its prevailing metric, the G-AP (Gap of Aggre…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 7.2

A bottom-up taxonomy of student discourse with a Socratic AI physics tutor

2026-08-07 · Syed Furqan Abbas Hashmi, N. Sanjay Rebello

General AI

Large language model (LLM) tutors are being deployed in introductory physics courses at a scale that produces transcript corpora far larger than traditional qualitative coding can absorb. A central question for physics education research (PER) is empirical and prior to any claim about effectiveness: what do students ac…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 7.2

GPTKB 2.0: Browsing, Querying, and Auditing a Disambiguated LLM-Derived Knowledge Base

2026-08-07 · Yujia Hu, Tuan-Phong Nguyen, Simon Razniewski

General AI

We present a web demo for exploring a large-scale disambiguated knowledge base (KB) materialized from a large language model (LLM). GPTKB 2.0 contains 38.4M triples over 1.6M canonical entities, together with 207.6K consolidated relations and 66K consolidated classes. Unlike prior LLM-derived knowledge bases that large…

Review
pending
Role
unreviewed
Read
soon
arxiv Score 7.2

Gaze Behavior in Visual World Experiments Can be Modeled With Off-the-shelf Language-Vision Encoders

2026-08-07 · Rahul Murali Shankar, Titus von der Malsburg, Sebastian Padó

General AI

The recent advances in neural language models have also spurred much work in computational psycholinguistics, asking whether neural LMs are also promising models of human language processing. However, work has been overwhelmingly focused on the unimodal case of written or spoken language. In contrast, multimodal experi…

Review
pending
Role
unreviewed
Read
soon
huggingface Score 6.4

When Privileged Guidance Misaligns: State-Matched Routing and Contextualized Self-Distillation for Multi-Turn Agents

2026-08-05 · Junzhuo Liu, Weiwei Li, Jun Ling, Peng Wang

General AI

Privileged on-policy distillation provides dense supervision for multi-turn agents by allowing a synchronized teacher to re-score the student's response at every turn with access to training-only references, such as successful trajectories. In interactive environments, however, the student's preceding actions continual…

Review
pending
Role
unreviewed
Read
later
huggingface Score 6.4

Characterizing the Quality Profile of AI-Generated C++ in Production

2026-08-06 · Michael Tran, Fred Lewis, Kun Yang, Saksham Thakur, Aditya Kini, Aditya Patil, Milad Hashemi, Parthasarathy Ranganathan

General AI

The widespread integration of AI coding assistants offers undeniable boosts to engineering velocity. Yet, recent studies point to a growing trade-off, revealing persistent challenges with code quality and maintainability. Industry leaders, including frontier AI labs, echo these concerns. As large language models are in…

Review
pending
Role
unreviewed
Read
later
arxiv Score 6.2

Beyond Post-Hoc Temperature Scaling: Bilevel Optimization for LLM Calibration

2026-08-07 · Ruochen Jin, Zhanliang Wang, Zongyu Dai, Jiancong Xiao, Bojian Hou

General AI

Preference alignment often makes large language models (LLMs) overconfident and poorly calibrated. Traditional post-hoc temperature scaling is inherently domain-dependent: a temperature fitted on one domain does not generalize across domains. This motivates us to modify model parameters during training to improve calib…

Review
pending
Role
unreviewed
Read
later
arxiv Score 6.2

Interaction Creates Dynamical AI Behavior Absent in Isolation

2026-08-07 · Bella Xinrui Li, Frank Yingjie Huo, Neil F Johnson

General AI

What will happen when AI agents interact in daily life, e.g. when one AI starts bossing another around? We find a counterintuitive answer that opens new avenues for out-of-equilibrium Physics. When a boss AI directs a stream of messages at the subordinate AI while ignoring its replies, it drives the subordinate into an…

Review
pending
Role
unreviewed
Read
later
arxiv Score 6.2

SetEasy: A Multi-Modal Classroom Engagement Assessment and Seating Optimization Framework

2026-08-07 · Zhihao Xie, Hongye Yang, Shien Liu

General AI

SetEasy optimizes classroom engagement in fixed seating grids. It fuses multimodal sensing (wristband physiology, 4K video, environmental data) and trains a v-Gage model grounded in a revised ISEQ. Each week, two-week engagement forecasts are mapped to a student-seat utility matrix, and CP-SAT generates seating plans u…

Review
pending
Role
unreviewed
Read
later
huggingface Score 5.4

Modular TTT: Rethinking Test-Time Training as Composable Modules

2026-08-07 · Bohao Tang, Zhen Qin, Yuqi Pan, Zheng Li, Pengfei Liu, Ya Zhang

General AI

Test-time training (TTT) views sequence modeling as an online learning problem in which fast weights are updated by an internal learning rule. Despite the growing number of TTT variants, existing approaches typically hard-code each variant separately, which makes it difficult to design new TTT methods and to isolate th…

Review
pending
Role
unreviewed
Read
later
arxiv Score 5.2

Bayesian Fair Division: Truthfulness in Picking Sequence with Correlated Valuations

2026-08-07 · Xiaolin Bu, Biaoshuai Tao

General AI

Sequential allocation mechanisms contain a class of widely studied mechanisms (e.g., round-robin) in the fair division of indivisible goods, where agents take turns picking items in a predefined picking order. It is known that the sequential allocation mechanisms are not truthful: when an agent's most preferred item is…

Review
pending
Role
unreviewed
Read
later
arxiv Score 5.2

SimWAM: A Simple World Action Model for End-to-End Autonomous Driving

2026-08-07 · Zongchuang Zhao, Xin Zhou, Tianyang Xu, Zhengyang Sun, Kaixuan Zhou, Honglin Li, Dingkang Liang, Xiang Bai

General AI

World-Action Models (WAMs) improve end-to-end autonomous driving by transferring video dynamics priors to action prediction, but existing methods require costly future generation at inference. We present SimWAM, a simple yet effective WAM that uses video generation purely as a training signal. It co-trains a pretrained…

Review
pending
Role
unreviewed
Read
later
arxiv Score 5.2

Unsupervised Adaptation of PDE Foundation Models

2026-08-07 · Ziye Song, Zhao Wei, Xin Yu, Ivor Tsang, Yueming Lyu

General AI

Pretrained partial differential equation (PDE) foundation models can generalize across different equations, but adapting them to unseen PDE systems typically requires dense solution data, which is often expensive or unavailable. To address this limitation, we propose an unsupervised PDE-based finetuning framework that …

Review
pending
Role
unreviewed
Read
later