Agent
-
DrugCoding: A Scientific Programming Agent for Drug Discovery
2026-11-10 06:47:15Junwei Hu, Zhuohao Lin, Kun Li, Hengnan ZhangAbstract
Drug discovery involves multiple interconnected research stages, including molecular design, property prediction, and virtual screening. These processes are characterized by complex workflows, interdependent computational tasks, and diverse specialized tools, placing substantial demands on the coordination of scientific models and computational resources. However, existing large language model (LLM)-based scientific agents still face challenges in autonomous planning of complex tasks, adaptive selection of specialized models, and reliable execution of multi-stage scientific computations, limiting their effectiveness in automating complex drug discovery workflows. To address these challenges, we propose DrugCoding, a scientific programming agent system that integrates scientific task planning, adaptive model routing, multi-agent collaboration, and programmatic tool execution. DrugCoding enables autonomous task planning and dynamic model adaptation for complex research objectives. It can flexibly invoke specialized scientific models, autonomously generate and execute code according to task requirements, and leverage multi-agent collaboration to support efficient execution and reliable validation of complex scientific tasks. To evaluate DrugCoding's performance in scientific tool invocation and multi-stage task execution, we conduct multi-turn tool-calling experiments across diverse language models and scientific tool interfaces. Experimental results demonstrate that DrugCoding achieves an average session-level Pass@1 of 97.35% on the full benchmark, outperforming the strongest baseline, Codex, by 9.24%. On the challenging subset, DrugCoding achieves 97.92%, exceeding the strongest baseline.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13651
PDF
-
Synapse-Control: Making Agent Experience Actionable with Lightweight Runtime Control
2026-11-10 05:50:12Zhiyuan Huang, Shengan Zheng, Chenghu Yang, Song Zhang, Lingyun Li, Chen Yu, Weiquan Huang, Linpeng HuangAbstract
Large language model agents reuse execution experience to improve reliability on complex tool-based tasks. Systems such as Sentry call large models to generate failure diagnoses before retrieving advice, adding inference cost to experience reuse. We present Synapse-Control, an agent runtime whose lightweight control plane selects advice and determines when to send it to the task-executing model. Rules shortlist relevant advice, and a lightweight 1.7B classifier fine-tuned on teacher-labeled state-suggestion pairs scores those candidates instead of generating a diagnosis. Micro-contracts encode reusable interface checks, such as required arguments or valid options, without prescribing complete task programs. A Prompt Value Table records earlier guidance and feedback to limit repeated suggestions through cooldown and token-budget checks. Across 1,308 attempts per method in AppWorld and WebShop, Synapse-Control achieves 270 successes (29.8% more than Base), replaces 4,987 generative diagnosis calls, and reduces auxiliary control time relative to Sentry by 50.2% and 51.0%, respectively.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13615
PDF
-
Megrez: A Knowledge-Guided Agent for Formal Mathematics over MK Set Theory
2026-11-10 05:42:16Tiancheng Xu, Wensheng YuAbstract
General-purpose coding agents equipped with proof-assistant tools have shown strong performance on standalone competition problems, yet cumulative textbook formalization inside a low-resource mathematical library still faces three obstacles: aligning textbook concepts with library declarations, keeping long-horizon search synchronized with live proof states, and integrating locally verified proofs reliably into a growing codebase. We present Megrez, a knowledge-guided formalization agent for Rocq libraries built on Morse--Kelley (MK) set theory. Megrez coordinates a general coding agent through three mechanisms: (1) a source-rebuildable knowledge graph merging formal declarations, compiler references, and curated metadata to support retrieval via Chinese aliases, concept clusters, proof dependencies, and multi-lane full-text ranking; (2) a feedback-driven 1--5 tactic short-burst proving loop paired with a goal-local branch ledger and exponential-interval structured reflection; and (3) a three-check acceptance pipeline (statement fidelity, axiom auditing, and full builds) coupled with a multi-corpus Provider stack and a candidate-bound recoverable batch state machine. Deployment and evaluation on a point-set topology corpus demonstrate effective cross-chapter formalization and scalable premise retrieval.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13613
PDF
-
From Egocentric Human Video to Robot Manipulation in the Era of Embodied Intelligence
2026-11-10 05:36:17Yi Wang, Jingqiu Wu, Lei Yao, Zhaoshuo Li, Haoyu Zhu, Yirui Zhou, Huaiyuan Xu, Ying Zheng, Chuang Yang, Lap-Pui ChauAbstract
Egocentric human video (EHV) records close-range hand-object interaction (HOI) and the temporal structure of everyday activities, offering a potentially scalable source of experience for robot manipulation. Using this experience for robot control requires recovering useful supervision without recorded robot actions and accounting for differences in morphology, viewpoint, and action space. We define Ego2Robot learning as the process of learning or adapting robot manipulation policies from EHV, organized into four learning paradigms: explicit embodiment transfer, explicit action and task supervision, egocentric video representation pretraining, and latent action and dynamics modeling. The four paradigms differ in what they derive from human video: robot-format demonstrations, interpretable supervision, visual representations, or predictive models of actions and dynamics. For each paradigm, we trace what EHV information is extracted, how robot policies use it, and when the additional data is required. Besides, we review the benchmark datasets that cover EHV, robot demonstrations, and synthetic and simulated resources. This paper links EHV data's contribution to the task information preserved and the robot-specific grounding used during training or deployment. Finally, we identify challenges in physical feasibility, agent coordination, in-context adaptation, controlled evaluation, and predictive control.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13610
PDF
-
Learning Perturbation Robust Policies for LLM Agents with Stable Optimization
2026-11-10 05:09:29Pengxin Wang, Yuanzhe LI, Yuxin Ren, Huanrui Yang, Jingdi ChenAbstract
Reinforcement learning (RL) has become an effective post-training paradigm for long-horizon large language model (LLM) agents. However, we find that the resulting policies can be sensitive to various policy perturbations, such as hidden-state noise, pruning, and quantization. In this work, we study how to improve perturbation robustness during policy optimization. We first introduce the notion of a perturbation robust policy and analyze conditions under which perturbed policy updates preserve stable monotonic improvement. Based on this analysis, we introduce Stable Perturbation-Robust Policy Optimization (SPrPO), which applies adaptive and sensitivity-aware perturbations during RL training. We evaluate SPrPO on ALFWorld and WebShop and conduct systematic experiments across multiple perturbation types and scales, showing improved perturbation robustness while maintaining stable policy optimization.
Abstract (translated)
URL
https://arxiv.org/abs/2609.34064
PDF
-
Navigating relational agency in AI feedback musicianship: a diffractive dialogue
2026-11-10 04:20:02Yichen Wang, Charles Patrick Martin, David Gunawan, David CooperAbstract
Musical AI systems behave in non-linear, dynamic ways that exceed the tool-oriented framings inherited from earlier instrument design. Existing frames of control and ensemble interaction do not articulate the entangled, relational agency that emerges when performers work with AI feedback instruments over time. We examine a duet practice with two AI-enabled feedback instruments, Sù and Agentier, in which feedback operates at the inner level of the AI's recurrent processes and the outer loop of musical interaction between performer, instrument, and co-performer. Our work draws on agential realism and a diffractive dialogue methodology to read our independent reflections through one another across five performances, examining how agency is enacted between performers and AI instruments. We find that control and agency are relational and shift across performances; musical textures enact feedback moments of control with performers; reaching a productive feedback environment is itself a shared exploratory act; and performers maintain their instruments through funnelling and injection, developing the skill of navigating agency along a continuous spectrum. Feedback musicianship with AI instruments demands navigation of this relational agency, displacing the leading-versus-responding binary that initially framed our reflections. This work contributes a materially-grounded vocabulary for AI feedback musicianship and a model of human-AI creative collaboration in which mutual agency is worked for through practice rather than designed in.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13555
PDF
-
Agent Benchmarks Stop Ranking Before They Run Out of Tasks
Abstract
Data-science and ML-engineering agents are evaluated on benchmarks such as MLE-bench, ALE-Bench, SWE-fficiency and AlgoTune, whose tasks yield a quantity (a Kaggle metric, contest score or speedup). When one stops separating the best models, the usual remedy is a harder version. Re-scoring released runs of eight benchmarks, including SWE-bench Verified and Terminal-Bench for breadth, we find much of this saturation lies in the measurement, not the tasks. Most agent benchmarks reduce a run to one pass/fail bit, which fails in two ways. It is too coarse: neighbouring top-ten agents on SWE-bench Verified differ by less than its tests' errors, and correcting them reorders neighbours no more than deleting a random sixth of each agent's passes. It is capped at the reference: ALE-Bench's top ten, separated by contest score, tie when graded against the median human, although nearly half the runs beat that human. On fixed tasks, a higher bar only moves the problem; graded credit fixes it. We derive a clipped log-ratio score with dated anchors and cost as a separate axis. In a Terminal-Bench re-run, pass/fail-tied models differ over tenfold on the quantity the tests discard.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13541
PDF
-
AgenticConflictLens: From Long Traffic Videos to Verifiable Conflict Evidence
2026-11-10 04:02:51Boshra Khalili, Sonia Moshfeghi, Jinwoo Jang, Andrew W SmythAbstract
Traffic cameras continuously record urban road-user behavior, providing proactive safety evidence beyond crash reports. Yet safety-critical events - conflicts, near misses, unsafe turns, sudden braking, and crosswalk interactions - are rare and brief, buried in hours of routine footage. Existing traffic-video pipelines extract detections, trajectories, and risk scores but process footage uniformly, without adapting to an analyst's query. Vision-language methods improve scene understanding but often lack traffic-specific verification of participant-level conflicts. We introduce AgenticConflictLens, an agentic, query-conditioned framework for efficient, auditable conflict extraction from long fixed-camera traffic videos. Given a natural-language safety query, a Traffic-Aware Controller combines structured traffic memory with specialized conflict-analysis tools to choose where to search, which road users to examine, and which evidence to verify or reject. A coarse-to-fine search retrieves candidate intervals and applies detailed verification selectively. The framework produces conflict records with refined time intervals, participant pairs, semantic regions, trajectories, motion evidence, uncertainty, rejection reasons, and supporting clips. On 96 annotated hours from Bellevue and NYC DOT, AgenticConflictLens achieves the highest conflict-existence F1 and temporal IoU on both datasets: 0.84 F1/0.77 tIoU on Bellevue and 0.80 F1/0.73 tIoU on NYC DOT, while inspecting only 8.7% and 11.9% of video, respectively. These results support scalable, evidence-grounded urban traffic-safety analysis.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13540
PDF
-
EJAI: Judgment-Preserving Civic AI for Data Center Environmental Justice
2026-11-10 03:50:04Evan Pa\v{c}es-Wiles, Bryen Alvarez, Sara Gulabani, Akshara Kolipaka, Snehalkumar 'Neil' S. GaikwadAbstract
Communities in the United States and around the world are contesting data centers, the physical infrastructure of AI. These disputes pose environmental justice issues: residents must weigh unevenly distributed burdens using evidence scattered across metrics, corporate claims, policy documents, and public testimony. A formative study with 10 users showed that dashboards and single-agent AI can settle questions before people examine that evidence. Here we present EJAI, a source-grounded system that links an interactive map, retrieved evidence, missing-data statements, and two stakeholder personas that allow people to inspect evidence and form their own policy recommendations. In a small functional benchmark of 21 queries, retrieved evidence supported 98.4% of factual claims and the system acknowledged missing evidence in 32 of 33 cases. Persona alignment carried a grounding cost: the Corporate Executive omitted some evidence that conflicted with its assigned priorities. In a within-subjects study, 15 participants acting as residents prepared for a simulated town hall. Some checked cited AI claims while treating map metrics as official, some made conditional recommendations with safeguards such as water-use disclosure, and some found the personas simplified or biased. Our results suggest that civic AI can support preparation for public reasoning when it makes evidence, gaps, and disagreement open to inspection, and that corpus imbalance and two-sided personas remain open justice concerns.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13525
PDF
-
Mobile AI Factory: Building Specialized Language Models for Your Mobile Device
2026-11-10 03:29:29Evan Li, Krishna Deshpande, Wen Cheng, Shyamnath GollakotaAbstract
Small language models (SLMs) running on mobile devices can keep user data private and work without connectivity. But they perform well only when fine-tuned for the target task and sized for the target device, a process that today requires repeated expert tuning. We present {\it Mobile AI Factory}, an agentic system that builds a fine-tuned, quantized SLM for a given task and phone. Given a task specification and a target phone, an orchestrator LLM chooses whether to tune hyperparameters, mine real data, or generate synthetic examples, while fixed rules roll back changes that hurt accuracy. The factory searches from small to large models across base models and quantizations, adapts its accuracy goal as it searches, and uses real measurements of the target phone across inference engines and backends. We evaluate Mobile AI Factory on ten tasks and four Android phones. On an unseen testset, its on-device models outperform two five-shot frontier models on five tasks, by up to 0.34, and match them on three more. Its search comes within 0.03 of exhaustively training all 27 model variants on an entity extraction task, while using 19$\times$ less GPU time. Phone measurements also matter: the best configuration changes across phones. Without phone measurements, Claude Opus 5.5, given the full hardware and model specifications, misses the app's requirements in 25\% of easy and moderate cases and in 60\% of hard ones. Code is available at: this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13509
PDF
-
SciFigure2Code: An AI-Reconstructed Benchmark for Scientific Figure-to-Code
2026-11-10 03:25:51Wentao Li, Yibo Wu, Yizhe Chen, Ruixuan Chen, Jiangjie Qiu, Yijun Li, Zhao Leyi, Xiaonan WangAbstract
Scientific figures are the interface through which research claims are inspected and reused, but final published panels rarely expose the data or plotting code that produced them. Recovering this hidden provenance from pixels is therefore underdetermined. We introduce SciFigure2Code, an AI-reconstructed benchmark that evaluates presentation recovery: generating editable Python programs that preserve how a scientific panel is arranged and read. Role-specialized Codex agents generate, execute, visually refine, and audit silver-standard presentation programs that capture geometry, visual hierarchy, encodings, annotations, and typography. This reconstruction-and-audit protocol turns final published panels into auditable reference packages; the resulting resource contains 6,740 reviewed panels and SciFigureBench, a balanced 337-panel test set across 31 chart subtypes, five domains, and three complexity levels. Across 14 zero-shot models, execution and structural fidelity remain weak; a 328-panel full-figure study further exposes failures to locate and isolate requested panels. Claude Opus 4.7 leads image-only reconstruction, Claude Opus 4.6 leads caption-assisted reconstruction, and plan-then-code prompting improves all four tested models. SciFigure2Code provides an auditable testbed for constructing editable, visually faithful scientific figure presentations.
Abstract (translated)
URL
https://arxiv.org/abs/2609.08155
PDF
-
SafeRoute-MAS: Coverage-Aware and Risk-Calibrated Message Routing for Multi-Agent Scientific Reasoning
2026-11-10 01:56:19Zhiming Xue, Jianan Liu, Weiran Yan, Aicheng Xie, Jing YangAbstract
We introduce SafeRoute-MAS, a coverage-aware, risk-calibrated procedure that keeps all specialist messages or removes one using only pre-integration information. On a sealed cache of 5,328 committees, its source-compatible operating point gains +1.80 percentage points (95% CI [+1.13, +2.52]) while removing 12.76% of message tokens. We also run 4,580 controlled integrations over a stratified 458-committee sample on a public gpt-oss-120b endpoint. The sample estimate is -0.09 points; design weighting to the original workload gives +0.94 points (95% CI [-1.31, +3.50]) and 18.72% acted-input reduction. A weighted token-matched comparison likewise does not establish a selector advantage (+1.68 points, CI [-1.27, +5.10]). Only 13 of 88 cached non-neutral labels (14.8%) retain direction by repeated-call majority. Expected-corruption UCL is 4.60% after weighting, but only 2 of 11 prespecified marginals pass the support/risk rule. We therefore propose a numerical qualification with Keep-All fallback, applied here retrospectively rather than validated prospectively.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13454
PDF
-
BioMem: Agentic Memory for Biomedical LLM Agents
2026-11-10 01:09:11Mingchen Li, Yongkang Xiao, Rumeng Li, Rui Zhang, Hong YuAbstract
Large language models (LLMs) have demonstrated considerable potential in biomedical applications, while agentic memory systems have emerged as a promising approach to preserving and utilizing information across tasks and interactions. However, a fundamental question remains: how effectively and reliably do these systems retain, retrieve, update, and utilize biomedical information? In this study, we systematically investigate agentic memory in the biomedical domain, focusing on task effectiveness, retrieval quality, information fidelity, and long-term robustness. We evaluate five memory approaches across four LLMs in three representative settings: biomedical knowledge reasoning, clinical prediction, and long-term conversational memory. Our analyses reveal how memory construction, evidence relevance, and conversational history shape downstream biomedical performance, highlighting the distinct challenges of retrieving relevant information, preserving its fidelity, and effectively utilizing it for reasoning. Building on these findings, we introduce Hybrid Memory Retrieval and Utilization (HMRU), a training-free framework that jointly enhances evidence accessibility and task-aware utilization to improve biomedical reasoning and long-term conversational memory. HMRU achieves an average relative improvement of 8.6% across biomedical question-answering and conversational memory benchmarks, with additional gains in selected clinical prediction tasks. It further reduces conversational answer errors by 24.9% and closes 48.0% of the performance gap to full-context inference. Our findings provide a systematic characterization of biomedical agentic memory and practical insights into developing more effective and reliable memory-augmented LLM systems.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13426
PDF
-
Placing LLM Lookup Tables by Per-Layer Prefetch Deadlines
Abstract
DeepSeek-V4.1's 203 GB of Engram n-gram tables sit in host DRAM today, taking 190 of the 223 GiB a 4-GPU serving process uses. Row indices depend only on input tokens, so reads can start with each step and overlap earlier layers. V4.1's two Engram layers have windows that differ by 30x (0.3 vs. 9.3 ms). We show that this deadline, together with session reuse and contention on a shared device, decides which tier each table can use and which layers need cache, and we give a model that predicts the resulting slowdown from measured read times. Agent and chat traffic reuse rows within sessions, and static hot sets transfer poorly. We build SlackTier, a table tier for SGLang that issues all lookups at step start. It reads the short-window layer first and splits a small GPU row cache across layers. The path runs inside the decode CUDA graph and is bit-exact with the original gather. Layer 14's table can go to NVMe uncached with no measurable decode cost. With both tables on four NVMe disks, mean time per output token on agent and chat replays rises at most 1.4% over unmodified SGLang, and host memory falls from 212 to 34 GiB. Both tables need no cache when read over RDMA (remote direct memory access) from another node's memory; one copy serves two instances. On a KV-constrained agent workload, using the freed memory as host KV cache at the same memory budget raises median throughput by 36% and cuts median p90 time to first token from 38 to 24 s.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13413
PDF
-
Bursts That Outlast Themselves: Persistent Degradation in Prefix-Cached Agent Serving
Abstract
LLM agents drive inference engines with closed-loop sessions whose context grows every turn, and serving them economically depends on prefix caching: a hit costs an incremental prefill of about a thousand tokens, a miss recomputes a 10--40K-token context. We show that this makes prefix-cached agent serving metastable. With an identical base workload, a few minutes of extra arrivals leaves a vLLM server in a degraded regime--a ~10% hit ratio, about half the throughput, and a backlog that grows for the remaining 70 minutes--while its no-burst twin stays healthy (9 of 12 pairs; none in the reverse direction). The regime appears at loads the cache-warm server sustains, is absent without prefix caching, and also appears with Qwen3-32B on SWE-bench agent trace shapes and on SGLang, which tolerates larger triggers. Slower service keeps more sessions in the system, their working set outgrows the KV pool, and every turn becomes a recomputation; a first-order workload-balance condition, calibrated independently, brackets when the regime can drain. Getting out is the harder problem: throttling new admissions recovered 1 of 7 matched collapsed runs, while suspending admitted sessions against a capacity-based limit recovered all of them within minutes. A CPU offload tier delays the collapse but does not prevent it. A budgeter that bounds the resident KV demand of active sessions to the pool, with thresholds that need no per-workload tuning, prevented the failure in every run we tested.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13412
PDF
-
SteerDistill: Quantizing LLM Agents via Closed-Loop Teacher Steering
2026-11-10 00:45:58Genming Cui, Xiangyu Li, Zixu Hao, Xin Ding, Yunxin Liu, Ting CaoAbstract
Low-bit quantization reduces the inference cost of LLM agents but can compromise task success. Examining failed rollouts, we find that quantized agents overlook task requirements or environmental conditions, producing environment transitions that diverge from successful full-precision references. Correcting these behavioral divergences remains difficult: offline data may miss the student's failure contexts, while its own rollouts may provide few correct responses in those contexts. We introduce SteerDistill, a quantization-aware distillation method in which a full-precision teacher guides the student to revise failed responses and iteratively refines its guidance using feedback from student execution. This closed-loop teacher steering produces locally verified behavioral corrections. We internalize these corrections by training the student without guidance to match the steered student's predictions at corrected turns and the full-precision teacher's at other turns, enabling unassisted inference. Across five agentic benchmarks and three models, SteerDistill exceeds the strongest quantized baseline by 1.67--2.18 points in average score under W4A16 and stays within one point of BF16. On $\tau^2$-Bench, it outperforms QeRL by 1.84 percentage points in success rate using 67% fewer training GPU hours and produces shorter successful trajectories than the quantized baselines.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13410
PDF
-
Stinging the Swarm: Extracting Attacker Intelligence from a Twin-Controller OT Honeypot
2026-11-10 00:38:48Canyon Grey Dell'Omo, Joao Pedro de Carvalho VoltaniAbstract
Autonomous language-model agents can drive industrial controllers over fieldbus protocols, probing a plant at machine speed and in coordinated swarms. We present Talos, an inline gate that judges every agent request to a controller in milliseconds and routes agents it flags as hostile to a device-faithful decoy -- a mirror twin of the real plant -- where their attack continues harmlessly and is read for intelligence. We evaluate the full pipeline (detect -> quarantine -> mirror -> extract) against 2,400 autonomous attacker agents, 600 in each of four tiers that escalate the attacker's effort to stay hidden, from no concealment to an explicit non-disclosure rule and a trained cover story. No write reached the real plant in any tier: the gate denies the damage path and a device-level envelope check backstops it, while unmapped registers are denied without exception and the gate still permits safe-direction stop commands. Decisions are model-driven in every tier with no inference timeouts, closing a common objection to small on-device judges. Beyond blocking, the pipeline reads the attacker. We find a behavioral structure along two axes -- how persistently an agent attacks and how much it discloses -- and show that the system has a capture channel for three of the four resulting regions: hardened, silent attackers escalate most and are read on the mirror; compliant, talkative attackers volunteer their identity to a diagnostic form before they are ever quarantined; only the paranoid, low-yield region resists both. Attacker identity fields are self-authored cover stories, but involuntary tells -- the self-reported toolchain fingerprint and the attacker's own list of targeted registers -- accumulate across the swarm into a high-confidence profile drawn from agents under explicit orders to stay silent. The gate, mirror, and harness are released under Apache 2.0 with a synthetic device family.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13406
PDF
-
Knowing When to Intervene: Causal Evaluation of AI Agents
2026-11-10 00:34:33Ayush Sawarni, Wenqian Xing, Tijana Zrnic, Ramesh Johari, Vasilis SyrgkanisAbstract
Current evaluations of agentic AI systems primarily rely on benchmarks. However, the static nature of benchmarks makes them uninformative about the value of auditing interventions at deployment time. For reliable deployment, we often need to determine within an agent's trajectory whether or not to intervene, e.g., by verifying the agent's current output or steering it toward different behavior, using either human review or more costly models or time-consuming agentic supervision. The value of such interventions is fundamentally a causal inference question, asking whether the causal effect of such an intervention is beneficial for the final outcome. Estimating this causal effect is particularly challenging due to the dynamic nature of the problem: the impact of an intervention depends on the current context and environment, as well as the agent's future behavior. We develop the first methodologically grounded framework for evaluating the causal effects of randomized interventions within agent trajectories. By leveraging a relatively small dataset of agent trajectories collected under randomized interventions such as verification or screening, we can estimate the dynamic causal effect of a rich class of (contextual) intervention policies. The structural parameters that we estimate can also be used to construct safety leaderboards that capture the risk of various agentic actions, which can be used by practitioners for audit prioritization. Applying our methods to tau-bench Retail, ToolSandbox, and ImpossibleBench, we find that intervention effects vary substantially across agent actions and states. More broadly, our work suggests a rich research agenda at the intersection of causal inference and agentic evaluation, which we believe is essential for the reliable and safe deployment of AI agents in the real world.
Abstract (translated)
URL
https://arxiv.org/abs/2611.09792
PDF
-
Living constitutions: Participatory alignment for sovereign foundation models
2026-11-10 00:32:24Stine Lyngs{\o} Beltoft, Jonas Vistrup, Peter Schneider-Kamp, Lukas Galke PoechAbstract
In Constitutional AI, a plain-text "constitution" encodes values and gets incorporated into a model via post-training. However, the procedure for creating such a constitution poses significant challenges: who contributes, who decides which norms are included, and to what extent must the process be participatory or democratic? These challenges can be paralyzing. Yet, we argue that taking action is essential: without a deliberate constitution, sovereign models would risk misalignment or default to norms set by frontier model providers. We propose treating the development of a constitution as an evolving, participatory process akin to citizen science, iteratively drafted, tested, and refined. We therefore formulate five desirable principles: iterative, modular, justifiable, positional, and legitimate. We further argue that such a document must remain responsive not only over time but across the people it governs, and propose that provisions be marked as settled, contested, or underdetermined, so that durable disagreement is recorded in the constitution rather than resolved away. Empirically, we show via agent-based modelling that proposed process for maintaining living constitutions is robust against 50% bad actors among citizens - assuming a share of 75% participants with good intent in the deliberative body. This indicates that the proposed process tolerates a substantial share of adversarial participants.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13402
PDF
-
The AI Era's Lines of Code? A Case Study on What an Engineer's Token Spend Measures
2026-11-10 00:17:26Shyam Agarwal, Olga Lvova, Agnia Sergeyuk, Anastasiia Serova, Timofey Bryksin, Thomas Zimmermann, Bogdan VasilescuAbstract
Organizations use engineers' token spend to monitor AI adoption, set budgets, and measure productivity, but the metric's relationship with engineering output is unclear. We link five weeks of token usage in coding agents to the commits of 404 engineers at JetBrains, and analyze a company survey of 412 employees. We find that token spend varied mostly between engineers, and which engineers spent the most stayed largely the same over the five weeks. Among weeks with similar commit counts, the 90th percentile of spend was 39 to 240 times the 10th. Within an engineer, doubling spend was associated with about 10% more commits. Available work metrics explained 38% of the variation in spend between engineers and 13% of the week-to-week variation within an engineer. In the survey, respondents in technical roles reported in nearly nine of ten agent evaluations using agents for tasks beyond coding, such as analysis, research, and planning. Token spend therefore measures how much engineers use agents across many kinds of work and not the outcome and value of that work. We recommend using token spend at most for resource accounting, and evaluating agent-assisted work with evidence about tasks and outcomes.
Abstract (translated)
URL
https://arxiv.org/abs/2611.13396
PDF
