The upcoming meeting, taking place on November 10 at Mila, will explore how we can collectively develop, govern, and deploy high-performing, reliable, and secure agentic systems by connecting academic researchers, industry experts, and practitioners.
Join us on the Venture Scientist Bootcamp, a full time, 4-month incubator at Mila, built specifically for deep tech founders with elite STEM backgrounds. You have until November 21, 2026 to apply!
We use cookies to analyze the browsing and usage of our website and to personalize your experience. You can disable these technologies at any time, but this may limit certain functionalities of the site. Read our Privacy Policy for more information.
Setting cookies
You can enable and disable the types of cookies you wish to accept. However certain choices you make could affect the services offered on our sites (e.g. suggestions, personalised ads, etc.).
Essential cookies
These cookies are necessary for the operation of the site and cannot be deactivated. (Still active)
Analytics cookies
Do you accept the use of cookies to measure the audience of our sites?
Multimedia Player
Do you accept the use of cookies to display and allow you to watch the video content hosted by our partners (YouTube, etc.)?
Publications
recAMPLIFY: Large-Scale Pretraining of Recursive Protein Language Models
Protein language models (pLMs) are powerful tools for protein annotation and mutation-effect prediction, but progress has largely depended o… (see more)n scaling model size and data, making them expensive and prone to memorization. Test-time scaling offers a complementary path, but methods developed for autoregressive LLMs do not transfer directly: masked encoders lack a generation loop, and amino acids do not verbalize in a reasoning language. Instead, we adapt recursive pretraining to masked pLMs, where a weight-tied Transformer core refines residue-aligned hidden states over an adjustable number of cycles. We instantiate this recipe as recAMPLIFY, a pLM with 155M unique parameters trained on the same 2T tokens as AMPLIFY-350M. Despite 56% fewer parameters, it improves CASP contact recovery and three out of four supervised tasks, notably raising fold accuracy by 22.5%. Because optimal depth for recAMPLIFY varies across tasks and assays, we then introduce Gated Mixture Recurrence (GMR), a framework that decodes each protein from a probability-weighted mixture of hidden states visited during recursion. GMR employs a learned residual gate at scalar, low-rank, or per-dimension resolution to control how much of each cycle's proposed update is accepted into the state. This unifies halting and gating into a single adaptive computation objective while recovering PonderNet-style halting as a special case without gates. Layered on top of recAMPLIFY, GMR further reduces expected layer depth by up to 37%, although the accuracy trade-off varies by task. Mechanistic analyses show that residual gating controls recurrence stability beyond the trained depth and that halt depth tracks biological properties.
As coding models and agent harnesses become more capable, reliable security verification is becoming a central bottleneck. Improving secure … (see more)code generation requires a signal whose reported progress remains aligned with actual program behavior under optimization; otherwise, an optimizer can improve the signal's verdict without removing the underlying vulnerability. Existing model-based approaches require models to either infer program behavior from source code or specify correct outcomes for adversarial tests, placing substantial reasoning demands on the security signal. We introduce \falcon{}, a counterexample-grounded verification framework that separates attack exploration from behavioral verification and connects them through execution traces of the program under test. In test-time repair, \falcon{} yields substantially stronger held-out functional-security gains than alternative security signals across secure code generation benchmarks, while preserving or improving functionality. As an RL reward, \falcon{} also outperforms static-analysis and learned rewards on held-out functional-security, even when all policies successfully optimize their respective training rewards. Combining \falcon{} with either proxy reward further improves held-out performance, suggesting that execution-grounded feedback can mitigate reward overoptimization while preserving useful guidance from existing security signals.
Abstract Tropical forest canopies host diverse tree and liana assemblages, yet key foliar and reproductive traits needed for taxonomic ident… (see more)ification of canopy plants are often difficult to see from the forest floor, creating a bottleneck in inventories. We introduce a drone‐based workflow that automatically acquires hundreds of close‐up, ultra‐high‐resolution photographs (~0.4 mm pixel −1 ) of selected tree crowns per day, enabling efficient botanical and ecological studies of canopy plants (demo: https://youtu.be/80goMEifpc4 ). The approach leverages a compact DJI Mavic 3 Enterprise (M3E) with wide‐angle and telephoto cameras to (i) generate a high‐resolution digital surface model and orthomosaic via structure‐from‐motion photogrammetry and (ii) capture close‐up photographs from ~6 m above the canopy at preplanned locations. Photographs are accurately (13 cm error) geolocated with real‐time kinematic (RTK) global navigation satellite system to support spatial co‐registration with photo
Recent time series foundation models (TSFMs) can consume covariates in-context, extending zero-shot forecasting to problems driven by extern… (see more)al factors. This enables scenario analysis: forecasting how a target responds to events that set covariates to chosen future values. Whether they can do this reliably is an open question. We introduce WhatIf-Bench, a benchmark measuring how well forecasters use covariates to forecast a target's response to future events. WhatIf-Bench is semi-synthetic: time series come from mechanistic simulators of data centers, power grids, road traffic, and water networks, with controlled covariate-target relationships grounded in each system. Each task presents a history of events intervening on covariates, then asks to forecast a future event that either replays seen values (Replay) or sets them to unseen ones (Novel). Varying the number of covariates, historical events, and event length yields 37,665 tasks. Evaluating TSFMs, frontier LLMs, and simple regression baselines, we find the LLMs reach the highest skill (improvement over a naive baseline), ahead of the best TSFMs. Every model forecasts well on Replay futures (48 to 85% skill) but drops to -7 to 50% on Novel futures, with the regression baselines falling the most. Skill on Novel futures also falls as events set more covariates jointly. More historical events raise skill on Novel futures across the board, but large differences remain. WhatIf-Bench reveals generalization to Novel futures as the core challenge, and offers a testbed toward arbitrary scenario analysis.
Health systems are deploying large language models (LLMs) into clinical workflows, often on the assumption that a second automated check can… (see more) only reduce error. But a second reasoner reduces error only insofar as it fails differently from the first. We say that a clinician and a model have complementary vulnerabilities to the extent that they break down through different mechanisms and on different cases. The two reasoners' failure substrates differ: they are embodied and cognitive in the clinician, and statistical and generative in the model - so their weak points need not line up. That divergence is plausible rather than guaranteed, and the independence it creates is fragile, eroding when clinicians defer to the model or the model conforms to the clinician. We therefore propose a predeployment audit in which each site measures error overlap on its own cases. We also propose a governance model with which the site keeps those errors independent over time. Health systems should then judge a clinical artificial intelligence (AI) tool by whether its failures overlap with their clinicians' failures, as well as by its accuracy.
Over-Defense in Coding Agents Begins at Silent Forks
Hanjun Luo
Xuecheng Liu
Xiucheng Zhang
Zhuoning Xu
Yingbin Jin
Yankai Chen
Xue Liu
Hanan Salam
Language models behind coding agents are post-trained with outcome rewards, and recent methods concentrate credit on the turns that decide t… (see more)he outcome. We find that defensive actions beyond what a task requires escape both. They begin at silent forks, the turns where every next action leads to the same outcome. Such actions leave the success rate unchanged, and frontier models take the disproportionate branch there in up to 91% of decisive actions under implicit budgets. The most direct fix, a checker in the trajectory reward, reaches these forks only through fully proportionate trajectories, leaving 74% of GRPO groups with identical rewards and behavior unchanged. This motivates a rethink of where learning signals belong, namely where behavior diverges, which a behavioral checker can locate and the outcome cannot. Our method, ForkDPO, contrasts a proportionate and a disproportionate action at the same prefix, at silent forks on both sides of the required behavior. On a test set of unseen repositories, it lowers a 4B model's disproportionate behavior from 70.3% to 58.9%, against 67.7% for DPO at randomly chosen turns, establishing behavioral divergence as a learning signal for how agents act beyond whether they finish. The code is publicly available at https://github.com/Astarojth/silent-forks.
Background: Mechanical and traumatic asphyxial deaths constitute an important category of unnatural deaths encountered in forensic practice.… (see more) Their epidemiological profile varies according to geographical, environmental
2026-10-17
International Journal of Drug Delivery Technology (published)
Training agents directly on the task they will be deployed on often leads to the best policy performance, but training on physical robots ca… (see more)n cause damage through falls, so minimizing training falls is critical. A standard mitigation hands control to a separate recovery policy whenever the agent leaves a designer-specified safe region. The resulting mixed-policy rollouts bias the on-policy update, and importance sampling cannot remove this bias when the recovery policy is deterministic. We propose SafeExplorer, a drop-in modification of proximal policy optimization (PPO) built on an unbiased policy-gradient estimator of the mixed-policy return. The estimator uses the score function only inside the safe region and never evaluates the recovery policy's density, so deterministic recovery policies pose no problem. On final reward, the estimator matches or outperforms truncated importance sampling, which is itself biased for that same gradient. SafeExplorer also speeds up learning near the safe-region boundary with a closed-form critic target at recovery-triggering states when dynamics and recovery are deterministic, and an imitation loss that copies recovery actions only from successful recoveries. On three simulated locomotion environments with ten seeds each, and given an already available recovery controller, SafeExplorer incurs
2026-10-17
Transactions on Machine Learning Research (accepted)
General-sum games are more challenging than fully cooperative or adversarial settings because agents may deceive or exploit one another. Ach… (see more)ieving high social welfare requires sustained cooperation, while a single defection can drive agents toward a welfare suboptimal Nash equilibrium. Moreover, assuming access to other agents' rewards or internal states is often unrealistic and implicitly introduces additional cooperation into the setting. To address this issue, we introduce \textit{Reciprocal Q-Augmentation} (RQA), a method that uses a latent-variable opponent model to augment the Q-function with estimates of how one's actions affects other agents. RQA allows agents to learn both cooperative and defective policies using only their own Q-function. We show that, in social dilemmas, agents that all follow the cooperative policy of RQA learn to cooperate faster and reach higher levels of cooperation than fully cooperative algorithms. Moreover, a single RQA agent that switches appropriately between its cooperative and defecting policies can induce cooperation in agents trained with standard RL algorithms, which would otherwise end up defecting. We evaluate our approach on the Iterated Prisoner's Dilemma, Iterated Traveler's Dilemma, Iterated Stag Hunt, and \texttt{CleanUp-Lite}. Our experiments show that the algorithms implementing RQA outperforms fully cooperative methods with shared rewards and enables cooperation with algorithms that otherwise fail to reach high social welfare.
Reward fine-tuning is an effective way to adapt pretrained diffusion and flow-matching models to downstream objectives, but aggressive optim… (see more)ization often leads to mode collapse and distributional drift. Existing methods attempt to control this drift with reverse-KL regularization, implemented as an
Value-at-Risk (VaR) is a widely adopted quantile-based risk measure in practice, yet incorporating VaR objectives into Markov decision proce… (see more)sses (VaR MDPs) remains a longstanding challenge. Recent work has derived dynamic programming equations for VaR MDPs in finite state and action spaces but left the case of more general spaces unresolved. This paper establishes, for the first time, the validity of these equations in continuous spaces, for both finite- and infinite-horizon settings. Such cases naturally arise in many real-world dynamic decision-making problems. As an application, we study an optimal execution problem with a VaR objective and demonstrate that our framework enables the characterization of optimal dynamic strategies, resolving an open question in the field.
Datamodels predict how a model's behavior changes when different parts of its training data are included or removed. However, linear datamod… (see more)els assign fixed contributions to training units and cannot represent non-additive joint effects. We introduce Tensor Network Datamodels (TNDMs), which use low-rank tensor networks to model dependencies among training units while keeping the representation compact. TNDMs learn a map from subsets of the training data to the outcome of training on those subsets, allowing them to represent interactions among several training units. We prove that complementary and redundant training groups can induce high-degree subset functions with compact low-rank representations. A single fitted TNDM can then be used for several training data questions, including predicting the effect of removing data, measuring the importance of individual training units, identifying interactions between training units, and selecting useful subsets of the training data. We evaluate TNDMs on controlled tasks designed to contain strong interactions, as well as standard fine-tuning settings for classification and reasoning. On a controlled reasoning task, TNDMs outperform every baseline while observing only half as many training subsets, and TNDM-TT reduces prediction NMSE by