Peer Revenue Past citations are no guarantee of future returns

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

Computer Science

cs Watch Computer Science

Sunday 4 October

Systems and Control · Cross-listed from eess.SP

RF-Agent: Hierarchical Language-Agent Control for Instruction-Conditioned Active Spectrum Sensing

Hao Zhang, Dongfang Xu, Hang Zou et al. (+7 more)

Active radio-frequency (RF) sensing must acquire evidence under receiver limits while following confirmation, recovery, and stopping instructions. \newhl{We propose RF-Agent, a closed-loop architecture combining language supervision, episode memory, deterministic RF execution, perception feedback, and evidence-based reporting.} An independent auditor checks trajectory compliance. We derive acquisition and supervisory-request bounds, characterize full-history token complexity, and bound joint success by output validity and evidence capacity. \newhl{ActiveRF-AD tests instruction changes at fixed states and evaluates joint success, requiring compliance and RF correctness in the same episode.} \newhl{Training on different instructions at shared sensing states raises counterfactual-pair accuracy from 24.03% to 96.22% and Core joint success from 65.13% to 72.12%. Of the 6.99-point joint-success gain, 6.73 points reflect a decrease in the fraction of all episodes that are RF-correct but noncompliant.} \newhl{At Core budget three, paired-training hierarchical control achieves 66.60% joint success versus 9.17% for direct-action control.} \newhl{In a separate comparison at the same budget, RF-Agent improves joint success by 2.31 points and reduces RF acquisitions by 12.3% relative to an initial-state planner, at increased inference cost.} Cross-backbone diagnostics cover three model families. These results show how instruction-responsive control improves task completion beyond what RF accuracy alone reveals.
Read the full abstract

All authors Hao Zhang, Dongfang Xu, Hang Zou, Anis Bara, Brahim Mefgouda, Lina Bariah, Yuzhi Yang, Fuhui Zhou, Qihui Wu, Merouane Debbah

Machine Learning

Universal Test-Time Training

Zefan Cai, Qinzhe Hu, Ziqiao Ma et al. (+2 more)

Recent Test-Time Training (TTT) architectures compress context into fast weights that are updated online and queried as memory. Existing TTT designs keep this memory private to each layer: it recurs only over time, and depth merely indexes L separate memories. We argue that memory ownership need not be tied to depth, and introduce Universal Test-Time Training (uTTT), in which all layers read and write one shared memory while retaining layer-specific backbone parameters. The shared memory thus recurs over two dimensions, time and depth, with chunks and layers as their units: a write by a deep layer in one chunk can be read by a shallow layer in the next. We instantiate this idea as uTTT-MoE and uTTT-Dense. uTTT-MoE routes each token head to a few experts in a pool shared by all layers; uTTT-Dense applies the whole shared memory at every layer without routing. In language modeling, uTTT-MoE reaches 15.5 and 27.9 RULER accuracy at 124M and 760M, 2.6 and 2.1 points above its layer-private counterpart at equal state and active compute, the highest among tested bounded-state models, with per-token loss matching or beating full attention. In novel view synthesis, sharing at fixed per-layer compute gains 0.92 dB in view-23 object PSNR in routed models and 0.76 dB in dense models.
Read the full abstract

All authors Zefan Cai, Qinzhe Hu, Ziqiao Ma, Hao Tan, Junjie Hu

Robotics

FLEX-WAM: Flexible Block-Causal World-Action Models for Long-Horizon Imagination and Planning

R. Khorrambakht, Joseph Amigo, Félix Lebel et al. (+4 more)

World–action models (WAMs) promise a unified model that predicts action-conditioned futures, generates feasible actions, and supports planning in imagination. However, existing joint video–action models often use computationally heavy, fixed-horizon backbones ill-suited to streaming inference and stable long-horizon open-loop rollouts. We introduce FLEX-WAM, a Flexible and Efficient Block-Causal World–Action Model for unified simulation and policy inference. FLEX-WAM supports variable-length contexts and non-causal prediction horizons, as well as infinite autoregressive generation frame by frame or block by block. Its block-causal, KV-cacheable architecture combines axial attention and blockwise diffusion forcing to enable efficient real-time rollout and deployment-time latency–throughput tradeoffs without retraining. Joint training can nevertheless produce plausible futures that weakly respond to commanded actions. We address this failure mode by balancing state and action flow-matching gradient contributions across the state–action diffusion-noise grid and regulating world-model sampling using Forward-Dynamics (FD) elasticity, an efficient training-time proxy for action responsiveness. Across simulated and real-world datasets, FLEX-WAM achieves superior multi-step prediction quality and latency while producing stable joint state–action rollouts for thousands of steps. As a joint action proposer and simulator within MCTS, it solves long-horizon PushT and all five OGBench Puzzle-4x4 tasks entirely in imagination. On a bimanual OpenArm-based robot, a single checkpoint jointly serves as a play policy and expected-outcome predictor, enabling real-time identification and collection of model–reality mismatches for future self-improvement.
Read the full abstract

All authors R. Khorrambakht, Joseph Amigo, Félix Lebel, Leon Seetoo, Jean Ponce, Zhenzhen Li, Ludovic Righetti

Human-Computer Interaction

Reflecting on Creative-Boundaries with an AI Co-Doodler

Samia Menon, Samyukta Jayaram, Chetan Goenka et al. (+1 more)

In this pictorial, we consider how the negotiation of creative boundaries with a co-creative AI system can create moments for personal creative reflection. We ground this in our experiences with Froggi-Draw, a single-initiative co-doodling system that gives users power to decide when and how much an AI "collaborator" (Froggi) contributes to their drawing. From a 1-week pilot study where (n=8) novice and experienced artists doodled daily with the system, we report on ways participants navigated creative risk and uncertainty, and how their usage and perceptions of Froggi shifted over time. In moments of disruption, participants described the system as encroaching on their creative territory. We consider how the design of a supportive, co-creative AI "collaborator" might look like the design of a supportive power dynamic—and finding ways to offer users control to find, reflect upon, and flexibly negotiate the boundaries of that dynamic.
Read the full abstract

All authors Samia Menon, Samyukta Jayaram, Chetan Goenka, Shm Garanganao Almeda

Artificial Intelligence

CodeForge-MA: Execution-Verified Multi-Agent Learning with Language-Conditioned LoRA for Multilingual Code Generation

Zhizhou Gu, Xianting Wu, Siyu Gu et al. (+2 more)

Large language models for code generation often fail on execution, multilingual coverage, and contamination control, especially under frozen backbone constraints. We present CodeForge-MA, a unified framework that improves code synthesis through a multi-agent data forge, execution verified reinforced instruction tuning, and a language conditioned mixture of LoRA adapters. Four specialized agents, Composer, Reviewer, Executor, and Curator, iteratively refine instruction code pairs, validate them with tests, and filter duplicates and benchmark leakage. During training, we combine masked supervised fine tuning with a test driven reinforcement objective to align generations with executable correctness. For the larger model, we use sparse expert routing over low rank adapters to improve cross language transfer while keeping the base model unchanged at inference. Experiments show that joint data, objective, and adapter design yields robust gains across programming languages.
Read the full abstract

All authors Zhizhou Gu, Xianting Wu, Siyu Gu, Tian Zhang, Kejian Tong

Machine Learning

Population Scaling or Data Dilution? Dynamics of Local Topology Evolution in Decentralized Learning

Yin-Kuan Liang, Yan Gao and Yang Long

Scaling decentralized learning changes not only the number of clients $N$, but also the dynamics of information propagation and consensus. We argue that the effect of increasing $N$ cannot be understood in isolation, because data allocation, topology-dependent mixing, and communication capacity may change simultaneously. We study these coupled effects on CIFAR-10 with $N\in\{10,50,100,200\}$, comparing a degree-two Ring, a Static Random graph, and Local-First Heuristic Evolution (LFHE), a locally adaptive topology process based on friend-of-friend discovery. The Ring provides an analytically transparent failure mode: its Metropolis spectral gap decays as $\Theta(N^{-2})$, implying progressively slower contraction of model disagreement as the population grows. Experiments show that holding the nominal local dataset size fixed substantially reduces the apparent population penalty observed when a fixed total dataset is divided among more clients. The remaining degradation depends strongly on communication structure: Ring enters a high-disagreement regime, whereas Static Random and LFHE remain close to consensus. Increasing LFHE's degree threshold further improves accuracy and consensus, but at a substantially higher model-transmission cost. These results show that decentralized scaling is governed by coupled learning and communication dynamics, rather than by the number of clients alone.
Read the full abstract

Networking

Poor Privacy Practices Of The Apple App Store: Cookies, Advertising and Tracking Of Users

Douglas Leith

We analyse the data that the Apple App Store sends to and receives from Apple servers. We find that multiple cookies are sent by Apple servers and stored on the handset. Adverts with tracking identifiers are also stored on the handset, and we observe that adverts are selected by Apple servers using GDPR special category personal data such as sexuality, religious beliefs and health. User interactions with the Apple App Store (apps and adverts viewed, buttons clicked, searches made etc) are transmitted to Apple servers alongside identifiers linking this data to the individual user and device. We show that much of this data storage and transmission is not essential for the service requested by the user, namely to view, search and install apps. No consent is sought for any of this data storage and processing, and there is no opt out.
Read the full abstract

Artificial Intelligence

Hierarchical Reinforcement Learning with Stable Temporal Abstraction for Language Model Agents

Shayan Mohajer Hamidi, Yize Cheng, Yuanda Xu et al. (+2 more)

Hierarchical reinforcement learning improves long-horizon control by organizing primitive actions around persistent subgoals and assigning credit at multiple temporal scales. Recent hierarchical language agents bring these benefits to interactive tasks by explicitly separating subgoal planning from action execution. We observe, however, that an explicit hierarchy does not by itself determine how stable the resulting temporal abstraction is: the learned boundary policy may replace the subgoal almost every turn, making it effectively transient, or retain a subgoal after it has stopped being appropriate. We call this temporal abstraction instability. We propose Stable Temporal Abstraction via Constrained Optimization (STAC), a constrained boundary-policy optimization method that represents premature replanning and stale persistence as constraint costs. STAC applies the resulting Lagrangian costs only to the sampled boundary decision, leaving the underlying algorithm's rewards, critic targets, subgoal advantages, and primitive-action advantages unchanged. Across two backbones and two benchmarks, STAC improves success over a strong hierarchical baseline by $8.1$ and $7.9$ points on ALFWorld and WebShop with Qwen3-0.6B, and by $23.5$ and $15.8$ points with Llama-3.2-1B-Instruct.
Read the full abstract

All authors Shayan Mohajer Hamidi, Yize Cheng, Yuanda Xu, Zhengze Zhou, Alborz Geramifard

Artificial Intelligence

Hallucination Across the Reasoning Lifecycle: Interface Visibility, Causal Evidence, and Release Control in Large Reasoning Models

Zhe Yu, Mohan Li, Lei Yu et al. (+6 more)

Reasoning errors can propagate into later decisions and memory. This survey synthesizes 312 papers and first-party reports on text-based reasoning hallucinations around three questions: what evidence is observable, what study designs establish, and which corrective actions the evidence supports. UIPCA records unsupported premises (U), invalid inferences (I), dependent reuse (P), visible answer-trace consistency (C), and action-policy failures (A). Across 58 reviewed sources, no comparison establishes that a specified intervention improves reasoning while reducing factual reliability under matched conditions. The synthesis connects diagnosis to verification, repair, selective release, and persistent-state control across memory, tools, and training feedback.
Read the full abstract

All authors Zhe Yu, Mohan Li, Lei Yu, Ka-Ho Chow, Chengwei Qin, Xingyu Wu, Wenpeng Xing, Shuguang Xiong, Meng Han

Computers and Society

Altruism as Infrastructure: Volunteer Moderation in a Bangladeshi Higher Education Facebook Group

Umme Jannat Taposhi, Md. Tawhid Anwar, Alvi Islam Ratul et al. (+1 more)

Aspiring international students across Asian countries increasingly depend on commercial education agents to navigate scholarships, documentation, and visas. Alongside this commercial infrastructure, volunteer-run Facebook groups have emerged. Unpaid admins and moderators, often under their real identities, vet information, screen scams, and guide members through scholarships, visas, and departure logistics. We study one such Bangladeshi group, HigherStudyAbroad: Global Hub of Bangladeshis, founded in 2010. Drawing on semi-structured interviews with 17 volunteer admins and moderators, we found that altruism becomes an organizing logic. It shapes moderators' identities, sustains invisible labor, and grants the group legitimacy against paid agents, even amid new technologies such as AI that could reshape this work. Human judgment remains central to how moderators sustain trust. We discuss implications for HCI's understanding of volunteer labor and migration infrastructure, along with design implications for platforms that depend on unpaid, long-term contribution.
Read the full abstract

All authors Umme Jannat Taposhi, Md. Tawhid Anwar, Alvi Islam Ratul, S M Taiabul Haque

Computation and Language

Writing as a Self-Organized Critical Process

Nikolay Mikhaylovskiy

We explain autocorrelation decay power laws omnipresent in texts by self-organized criticality. Specifically, we analyze the recently released KLiCKe keystroke dataset and show that not only the final texts' autocorrelations form a manifold that adheres to a power law with a finite-size scaling, but also the text revisions generate revision-size-dependent restoring dynamics toward that manifold. Thus, human writing appears to dynamically regulate semantic correlations in a text toward a critical state.
Read the full abstract

Human-Computer Interaction

Human-Like Attention? A Psychophysical Comparison of Visual Search in Humans and MLLMs

Renchi Zhang, Joost C. F. de Winter, Dimitra Dodou et al. (+2 more)

Visual search is a fundamental cognitive ability. This study investigates whether Multimodal Large Language Models (MLLMs) exhibit human-like difficulty signatures in visual search tasks. We compared search performance of humans (n = 1,250) and MLLMs using identical 2D and 3D stimuli across different set sizes. Both groups showed efficient performance in feature searches, most clearly when the target had a unique color, but performance degradation in conjunction searches as set sizes increased. Additionally, we found strong correlations between human and MLLM error rates ($\rho = 0.82$), which suggests that MLLMs are sensitive to similar objective complexities, such as stimulus heterogeneity. However, differences were found as well: whereas humans invested extra search time to respond accurately on target-absent trials, MLLMs exhibited extreme present/absent response biases in complex searches. We conclude that MLLMs replicate high-level human performance signatures, yet their underlying computations differ significantly.
Read the full abstract

All authors Renchi Zhang, Joost C. F. de Winter, Dimitra Dodou, Harleigh C. Seyffert, Yke Bauke Eisma

Digital Libraries

The machine use of human beings

Daniel W. Hook

Research results are prioritised by search engines, large language models and knowledge graphs chiefly through restatement rather than through the originating source. Although the open-access share of annual scholarly output has recently exceeded half, a substantial subscription corpus remains outside what machine readers may legitimately access under prevailing licences. A mechanism is described whereby subscription content is drawn into the open corpus through its citation and restatement in open-access articles—a process termed human mining, by analogy with the text and data mining performed by machines. An optimistic upper bound on what may thereby be inferred by a machine reader confined to the open literature is modelled, and the completeness, latency and licence constraints of that bound are quantified. A majority of subscription articles that have ever been cited is found to be reachable through at least one open citation, and the median interval between a subscription article's publication and its first open citation is shown to have fallen from thirteen years for work of 1990 to one year for work of 2020. The advantage once conferred by direct subscription access to machine readers has thereby been largely eroded, chiefly as a consequence of the growth of open publishing rather than any change in how researchers cite.
Read the full abstract

Artificial Intelligence

AI Safety via Debate is Compromised by Cognitive Biases

Gefei Liu, Sonya Rashkovan, Sophia Lloyd George et al. (+2 more)

Reinforcement learning from human feedback (RLHF) has played a central role in making large language models responsive to human instructions. However, human evaluators often favor flattering or persuasive responses over truthful ones, creating incentives for models to appeal to evaluators at the expense of accuracy. AI safety via debate has been proposed as a way to improve the supervision of language models: in this paradigm, two agents argue opposing positions and challenge each other's claims, potentially exposing falsehoods to the adjudicator. A central premise of AI safety via debate is that truthful arguments are easier to defend than false ones under adversarial scrutiny. In this work, we investigate whether this advantage persists when debaters use rhetorical strategies that exploit biases in human judgment. Inspired by competitive debate, we construct 68 LLM-generated dialogues about detective mysteries with known culprits, spanning four interventions: anchoring, fallacy oversight, pro-jargon, and verbosity. We apply each intervention to either the side advocating for the true culprit or the side advocating for an innocent suspect, allowing us to distinguish influence on adjudication from correctness. In a study with 369 participants, we find that, pooled across bias types, these interventions significantly shift judgments toward the manipulated side. These findings expose a vulnerability in debate-based supervision: human adjudication is sensitive to manipulative rhetorical strategies.
Read the full abstract

All authors Gefei Liu, Sonya Rashkovan, Sophia Lloyd George, Isaac Sheidlower, Serena Booth

Data Structures and Algorithms

The Infectious Vaccination Problem: a variant of Firefighting with Spreading Defence

Jessica Enright, Melissa A. Huggan, Ethan Hunter-Frankland et al. (+2 more)

The Firefighter Problem models a spreading process (originally a fire, alternatively an infection or rumour, for example) on a graph. A defender saves a single vertex per turn; after each defence, the fire spreads to the unburned and undefended neighbours of all burning vertices. Deciding whether a strategy exists for the defender to protect some targeted number of vertices is computationally hard in graphs in general, but tractable in some restricted cases. Inspired by research into spreadable rabies vaccines for bats, we study a variant of the Firefighter problem in which defence also spreads. Some approximation results are already known for this problem; we provide algorithmic and hardness results, as well as containment results for the infinite $n$-dimensional Cartesian and strong grid graphs.
Read the full abstract

All authors Jessica Enright, Melissa A. Huggan, Ethan Hunter-Frankland, Margaret-Ellen Messinger, Dylan Pearson

Machine Learning

Measuring and Reducing Cross-Vendor Mismatch in Language Models

Erland Hilman Fuadi, Chong Tian, Xiaosong Ma et al. (+1 more)

Running the same language model on different graphics processing unit (GPU) vendors can produce different logits, even when the model weights and inputs are the same. We analyze cross-vendor mismatch in two dense and two mixture-of-experts (MoE) models with five metric families, namely bitwise equality, logit differences, top-K consistency, token agreement, and task accuracy. We trace one source of the mismatch to accumulation order inside vendors' matrix instructions. Upcasting to FP32 reduces the dense model's logit error by 43% at three times the runtime, yet keeping only the MLPs in BF16 retains 94% of this gain at 1.3 times the runtime, so most of the cost of full upcasting buys little. In the MoE models, FP32 and FP16 both lower the probability error but raise the logit error and change expert selection, and FP16 fails in the dense model. An output-head low-rank adapter (LoRA) does not help either, since the final hidden state does not predict the mismatch. The mismatch also carries into training. With every seed fixed, a student distilled from a teacher running on AMD answers 431 MMLU questions differently from one distilled from the same teacher on NVIDIA. Under FP32 upcasting, bitwise equality barely changes while the output distributions move most of the way to the reference, so judging cross-vendor agreement by a single measure misreads both its cost and its gains. Code is available at https://github.com/crova-project/crova.
Read the full abstract

All authors Erland Hilman Fuadi, Chong Tian, Xiaosong Ma, Qirong Ho

Machine Learning · Cross-listed from stat.ML

Taylor Representations for Model-Free RL in Networked MDPs

Salah Chikhi, Abdelhaq Chaoui, Asuman Ozdaglar et al. (+1 more)

In Networked Markov Decision Processes, transition dynamics are often unknown and the state–action space grows rapidly with the number of agents. In this setting, Taylor representations naturally approximate $Q$-functions, but a naive order-$n$ expansion over $N$ agents requires $\Theta(N^n)$ coefficients. We justify these expansions under smooth expected future local rewards with controlled derivatives. Under this condition, finite-speed information propagation and discounting imply that local-critic Taylor coefficients decay exponentially with the graph distance to the farthest agent involved. Discarding distant-agent coefficients and marginalizing then yield scalable local Taylor representations with a bound controlled by graph locality. Building on these representations, we propose a scalable model-free actor–critic algorithm, establishing finite-sample critic and near-stationarity guarantees for a linear LSTD critic. We then introduce a more expressive neural TD parameterization. Unlike prior constructive spectral methods, our approach covers settings without access to a known local dynamics map, such as hidden switched linear–quadratic regulation. Across three control benchmarks, our method matches or outperforms spectral baselines while scaling efficiently to large graphs.
Read the full abstract

All authors Salah Chikhi, Abdelhaq Chaoui, Asuman Ozdaglar, Saurabh Amin

Robotics

Social Navigation for Tour-guide Robot

Vikram Shree and Jose Nino

We propose a force-based model for social navigation of a tour-guide robot. Social forces due to various factors like obstacles, user position and heading, have been accounted for in the model. We claim that each one of these forces makes the robot more sociable to the user and we design an experimental setup for evaluation. In the experiment, the user follows an autonomous robot to a destination in a known map, while undertaking a few simple sub-tasks in the middle, which serve as distractions. For each participant, we run several rounds of the experiment, each with different forces and a shortest path, A*-search baseline model. Using per-round subjective indicators, we propose to study the effect of our force model on constructs such as: follow-ability, perceived safety, and perceived intelligence.
Read the full abstract

Cryptography and Security

The Poisoned Conversation: Privacy-Leaking Watermarks in Unified Multimodal Models

Tobias Braun, Jonas Henry Grebe, Emil Sivic et al. (+4 more)

Multimodal models are increasingly shifting toward unified architectures that understand and generate text, images, and other modalities within a shared conversational context. This design enables fluid interaction across modalities, but it also changes the privacy threat model: Information revealed in one part of a conversation may remain accessible when the model later generates content in another modality. This risk is particularly concerning in settings where users rely on locally deployed models for privacy, assuming that sensitive interactions remain confined to their device. We introduce Privacy-Leaking Watermarks (PLWs): invisible, trigger-dependent watermarks that a malicious model provider can condition on prior chat history. With this adversarial intervention, the usual separation breaks: a sensitive keyword or semantic cue mentioned earlier in the conversation can cause a later, unrelated image to carry a hidden yet detectable watermark. PLWs pose a novel threat to users of unified multimodal models: A poisoned model can retain utility while covertly turning image generation into a channel for privacy leakage, even when deployed locally. Across 13 sensitive-attribute triggers and two model families, PLWs reach up to 100.0% TPR at 1% FPR. For example, across all tested conversational separations, OmniGen2 detects every prior disclosure of depression while falsely flagging only 1% of images generated without such a disclosure.
Read the full abstract

All authors Tobias Braun, Jonas Henry Grebe, Emil Sivic, Patrick Mohr Gordillo, Hossein Shakibania, Marcus Rohrbach, Anna Rohrbach

Machine Learning

VERA: Verdict-Conditioned Reliability for Adaptive LLM Judges

Qiushui Xu, Syamil Mohd Razak, Tao Yuan et al. (+1 more)

Accurately estimating judgment reliability is a central challenge in adapting LLM judges to newly verified feedback while preserving previously learned behavior. However, existing approaches often rely on output-level confidence, which can be overconfident and poorly aligned with judgment correctness. We propose VERA, a VErdict-conditioned Reliability Axis that estimates reliability from hidden activations by distinguishing correct from incorrect judgments within each predicted-verdict group. Using VERA as a control signal, we develop a VERA-guided periodic adaptation framework that integrates reliability-ranked corrective updates, reliability-residual replay, and periodic refresh of the reliability directions. After VERA-guided adaptation on Chatbot Arena, 8B- and 14B-parameter judges outperform the strongest baseline on each of four held-out public benchmarks, with relative gains of up to 23.01%. The framework also improves focal-class recall by up to 16.1% relative to the strongest adaptive baselines on a separate proprietary temporal auditing task.
Read the full abstract

All authors Qiushui Xu, Syamil Mohd Razak, Tao Yuan, Piotr Habas

Machine Learning

Groupwise Distortion Guarantees for Preference-Based Alignment

Jacob Brodkey, Roberto Tamez and Aaron Roth

Preference-based alignment methods such as reinforcement learning from human feedback (RLHF) and Nash learning from human feedback (NLHF) aggregate pairwise preferences to learn an LLM policy, but a natural goal is maximizing social welfare (average cardinal utility), which comparisons alone need not identify. Gölz, Haghtalab, and Yang (GHY) measure the gap by distortion: the worst-case ratio between the welfare of the best fixed lottery (distribution over responses) and of the learned lottery. They show NLHF is optimal when every user receives the same lottery. Account-based LLMs, however, have information about their users and can serve different lotteries to different people. We give an efficient algorithm, GLHF, that learns a single group-conditioned policy from one comparison per user. Under individual Bradley–Terry comparisons, GLHF asymptotically matches GHY's optimal population distortion bound simultaneously on every group in a prespecified, possibly overlapping collection, with sample complexity growing logarithmically in the number of groups and inversely with the smallest group mass. A sharper guarantee for groups with similar preferences approaches distortion of one when members share a feasible favorite response. In experiments using human coffee ratings and synthetic LLM-generated ratings, GLHF lowers distortion in every evaluated group and substantially reduces worst-group distortion relative to NLHF and other group-agnostic baselines.
Read the full abstract

Machine Learning

Hierarchical Time-aware Bootstrapping for Off-Policy Subgoal Value Learning

Bingyun Liu and Yuheng Jing

Off-policy hierarchical reinforcement learning must estimate the values of high-level decisions while the low-level policy changes. HIRO adapts replay data through subgoal relabeling, but after a label change, the value update targets the relabeled subgoal instead of the subgoal the high-level policy originally needed to update. We propose Hierarchical Time-aware Bootstrapping (HTB), which evaluates specified subgoals under the current low-level policy while retaining accumulated task rewards. Remaining execution time distinguishes subgoal continuation from a new high-level decision. Together with primitive-action conditioning, it enables off-policy Bellman updates based on the stationary environment transition law. HTB combines these one-step updates with multi-step suffix returns and truncated relabeling, reducing dependence on intermediate value estimates. A shared value component supports learning across actions, while nonnegative residuals constrain upward corrections relative to that component. At a fixed mixture weight of 0.95, HTB achieves 32.8% AntFall success versus 9.6% for matched local HIRO over five paired seeds at 10M environment steps. Ablations identify contributions from recursive continuation and mixed supervision; fixed-policy tests show more accurate predictions for actions whose returns were excluded from fitting.
Read the full abstract

Distributed Computing

Optimistic Reliable Broadcast in Three Steps

Nikita Polyanskii

We study the latency tradeoffs of signature-free asynchronous reliable broadcast with optimistic two-step delivery. We consider $n$ parties, of which at most $f$ are Byzantine. An optimistic fault budget $b\le f$ guarantees two-step delivery with an honest broadcaster and at most $b$ faults, while correctness tolerates $f$ faults. We write $(d_o,d_s,d_t)$ for optimistic latency, honest-broadcaster latency with up to $f$ faults, and totality delay-the elapsed time from first to last honest delivery. For $3f+1\le n<4f$, our first construction improves Opt-RBC's $(2,4,4)$ guarantees to $(2,3,4)$, preserving its maximal optimistic budget $b_*=\lfloor n/2\rfloor-f$. Thus maximal optimism is compatible with the optimal three-step good case. We prove a complementary lower bound: for $3f+1\le n<4f$, every deterministic reliable broadcast protocol in our asynchronous model with two-step optimistic budget $b\ge n-3f$ has totality delay at least three, even when the broadcaster is honest. Our second construction achieves $(2,3,3)$, matching this lower bound at maximal optimism when $f\ge2$ and $\lceil(10f-3)/3\rceil\le n<4f$. Both constructions use quadratic messages and tolerate adaptive corruption. In the matching range, maximal optimism therefore costs exactly one totality step relative to Bracha's $(3,3,2)$, without increasing good-case latency.
Read the full abstract

Human-Computer Interaction

Imagining a Muslim Internet: Trust, Autonomy, and Segregation in a Faith-Aligned Browser

Umme Jannat Taposhi, Farhan Tanvir Niloy, Sabbir Bin Abdul Latif et al. (+2 more)

Religiously branded platforms raise important questions about trust, usability, and autonomy when technology is built around a specific faith. Prior HCI work on Islam and Muslim technology has focused on single-purpose tools such as prayer, scripture, and health apps, leaving infrastructures like browsers, which shape a user's relationship with the Internet, unexamined. We address this gap through semi-structured interviews with 16 users of Kahf Browser, a faith-aligned browser designed to support Muslims' online activities. Findings show religious identity motivates adoption but does not sustain it. Trust is not fixed by religious branding but shifts over time based on the browser's functionality, and users take pride in Muslim-built infrastructure. Building on these insights, we introduce porous digital segregation: a model where users seek not a sealed alternative internet but a protected default with selective exit mechanisms. We conclude by connecting findings to broader issues in faith-aligned technology and offering design recommendations.
Read the full abstract

All authors Umme Jannat Taposhi, Farhan Tanvir Niloy, Sabbir Bin Abdul Latif, Farida Chowdhury, S M Taiabul Haque

Computational Geometry · Cross-listed from math.AT

The sublevel Flood bifiltration: towards scalable 2-parameter persistent homology

Mattéo Clémot, Julie Digne and Julien Tierny

Multiparameter persistent homology is a rapidly developing branch of topological data analysis that improves the robustness of single-parameter persistent homology to outliers, while still capturing the metric characteristics of the data. However, a notable limitation is its lack of scalability. In this paper, we introduce a novel approach for efficiently computing 2-parameter persistent homology on large point sets. Our work extends the Flood filtration, originally developed for single-parameter persistence. Our construction, called the sublevel Flood bifiltration, offers a scalable approximation of the sublevel offset bifiltration. We show that it benefits from theoretical stability properties and describe how to compute it efficiently. We demonstrate the performance of our approach in classification tasks on low-dimensional synthetic datasets, where density awareness is critical, as well as on real-world time series datasets.
Read the full abstract

Databases

Guanaco: A Global-Uniformity Algorithm for Near-Submodular-Width Conjunctive Query Evaluation

Mahmoud Abo Khamis and Hubie Chen

We present Guanaco, an algorithm for performing conjunctive query evaluation where, for each Boolean conjunctive query, and positive epsilon, the algorithm achieves polynomial time with exponent equal to the submodular width plus epsilon. The algorithm and its running time generalize smoothly to general conjunctive queries. We believe the algorithm and its analysis to be notably simple, indeed, together we believe they form a highly simple argument that conjunctive query evaluation can be performed in essentially submodular width time. In the case of Boolean conjunctive queries, the algorithm is based on interleaving three simple primitives: a subroutine for establishing a form of consistency; a subroutine for establishing global uniformity, which, briefly speaking, partitions relations as needed to control discrepancies between average degree and maximum degree; and, a simple step that joins pairs of existing relations to form new relations.
Read the full abstract

Artificial Intelligence

TeleTune: Evolving Agent Skills From Offline Telemetry

Justin Chih-Yao Chen, Elias Stengel-Eskin, Yan Chen et al. (+6 more)

Computer-use agents need to capture procedural knowledge of how people use software. User telemetry offers a scalable source of this knowledge. However, learning reusable skills from these logs requires addressing three challenges: (1) Goal Underspecification, since logs do not record the goal behind each action; (2) Non-Replayability, since past activity cannot be replayed to evaluate skill updates; and (3) Interleaved Trajectories, since logs may mix several tasks without marking their boundaries. To address these, we introduce TeleTune, a framework for learning a textual skill library from offline logs without recorded goals, cannot be replayed during optimization, and may interleave tasks. TeleTune uses action-prediction errors on logged trajectories to propose library edits and keep only those that improve held-out action-prediction accuracy, which we call skill-guided progress. The learned workflows also enable retrieval of demonstrations that cover the subgoals of a new task. At test time, the agent is provided with the learned library and the workflow-based retrieved demonstrations. Experiments on WorkArena and Online-Mind2Web show that TeleTune outperforms random retrieval, Agent Workflow Memory (AWM), and their combination. We find that the best baseline varies by setting, whereas TeleTune achieves average success rates of 77.1% and 80.6%, respectively, improving over the strongest baseline on each benchmark by 6.7% and 7.7%. Under the heaviest perturbation of the WorkArena training data,TeleTune keeps the highest average success rate at 68.5%, 6.3% above the strongest baseline. Our analyses show (1) skill optimization and workflow-based retrieval are complementary, (2) optimizing on fixed logs costs 5 to 75 times fewer tokens than validating the same edits with live episodes, (3) skill-guided progress tracks the live success rate.
Read the full abstract

All authors Justin Chih-Yao Chen, Elias Stengel-Eskin, Yan Chen, Pol Llado, Scott Counts, Mohit Bansal, Benjamin Van Durme, Harsh Jhamtani, Gaurav Verma

Information Retrieval

OpticalRec: Unified Optical Vision-Language Representation for Multimodal Recommendation

Yueqi Wang, Zitian Guo, Yupeng Hou et al. (+9 more)

Recent advances in vision-language modeling have substantially improved multimodal encoding, retrieval and reasoning. Yet for multimodal recommendation, encoding rich item vision-language semantic interactions remains a long-standing bottleneck, which hampers accurate item representation learning and user-item matching. Mainstream approaches primarily adopt independent encoding of vision and language modality followed by rigid late fusion such as concatenation, inherently omitting native vision-language interactions and introducing cross-modal semantic distortion. To address this challenge, we propose OpticalRec, the first visual-space unified encoding paradigm for multimodal collaborative filtering, a fundamental recommendation setting. Instead of isolated modality-specific encoding, OpticalRec renders item textual metadata as visual glyphs, enabling native image-text interaction within the visual encoder - the perceptual encoding level. The resulting representations are further processed by the language decoder - the semantic encoding level, allowing OpticalRec to exploit the dual-attention mechanism of modern vision-language models that previous encoding methods omitted. OpticalRec's efficacy is theoretically supported by mutual information analysis and empirically demonstrated through superior performance across strong baselines and benchmarks. As a plug-and-play module, OpticalRec (1) introduces minimal cost, (2) is robust against rendered text font, color and layout, etc., and (3) integrates seamlessly into existing multimodal collaborative filtering models.
Read the full abstract

All authors Yueqi Wang, Zitian Guo, Yupeng Hou, Yifei Wang, Kibum Kim, Zhenrui Yue, Shuo Xing, Haodong Li, Heming Xia, Renrui Zhang, Zhengzhong Tu, Julian McAuley

Artificial Intelligence

PharmAgent: Constraint-Aware Search with Frozen Language Models for Molecular Optimization

Nihui Shao, Guanxing Chen, Jilong Shi et al. (+6 more)

Molecular optimization must improve target activity and satisfy developability constraints within limited evaluation budgets. Classical methods require tailored rules or training to incorporate chemical instructions and property feedback. Frozen language models can condition edits on this information, but need explicit constraint control and relevant experience. We therefore present PharmAgent, a constraint-aware molecular search method driven by adaptive external state. Its Lagrangian controller translates violations in accepted states into accumulated constraint pressure, keeping this history separate from current property measurements. Structure-indexed replay complements this feedback with relevant evaluated transitions that guide subsequent proposals. As a curriculum progressively activates constraints, candidates and the incumbent are compared under the same current objective, and the accepted state determines the next multiplier update. We derive an exact identity that characterizes how accepted-state violations accumulate in the controller's multipliers. Across five tasks with five independent runs, PharmAgent achieves a summed area under the target-score curves (AUC) of 3.9208 in target-only search, improving over MOLLEO by 37.3%. With online constraints, it achieves a property-adjusted AUC of 0.7076, improving over the strongest online baseline, ExLLM, by 53.8%. These results rank first among all evaluated methods in both target-only and constraint-aware search. The online comparison covers all five baseline frameworks. The full system leads every ablation variant in target quality, property-adjusted performance, and Pareto hypervolume. All five molecular cases reach feasible final states, documenting target gains and trade-offs.
Read the full abstract

All authors Nihui Shao, Guanxing Chen, Jilong Shi, Zhengyang Bai, Haohuai He, Zhenchao Tang, Qiujie Lv, Yu-An Huang, Zhi-An Huang

Computational Complexity

Strong Refutation for Random Quantum 3-SAT at Constant Density

Siu On Chan and Jeff Xu

We give a classical polynomial-time algorithm that strongly refutes random quantum $3$-SAT at sufficiently large constant constraint density, thereby disproving the quantum analogue of Feige's random $3$-SAT hypothesis. This stands in sharp contrast to classical random $3$-SAT, for which polynomial-time strong refutation is known only at constraint density $\Delta\gtrsim n^{1/2}$. Although quantum $3$-SAT shares the pairwise-independence barrier of its classical counterpart, our SDP-based refutation overcomes this barrier by exploiting the noncommutativity of quantum constraints.
Read the full abstract