Dane Malenfant bio photo

Email CV

Twitter

LinkedIn

Instagram

Google Scholar

BlueSky

Publications

Research programmes

Questions that connect the papers

My work spans several publication venues, but it is organized here by the research problems that connect it: cooperation among learning agents, credit and generalization across time and structure, and the wider social and cultural contexts of AI.

Research programme · Ongoing

Reciprocity & Multi-Agent Learning

How can adaptive agents learn to cooperate when benefits are delayed, hidden, costly to produce, or mediated by another learner?

Variance in collective success across training for four policy-gradient variants
Accepted Code available

RLC 2026 · Montréal, Canada · August 2026

The Challenge of Hidden Gifts in Multi-Agent Reinforcement Learning

Dane Malenfant and Blake Aaron Richards

Question
How can agents learn reciprocal behaviour when helpful actions produce delayed or difficult-to-observe benefits?
Contribution
We introduce a shared-key grid-world that makes helpful actions unobservable, show that several state-of-the-art multi-agent reinforcement learning algorithms fail to obtain the collective reward, and derive a learning-aware policy-gradient correction that reduces variance and improves reliable convergence to collective success.
GPT-5.6 Sol query, team-success, and local-reward rates across query costs
Preprint Workshop paper

Social Simulation with LLMs @ COLM 2026 · San Francisco, USA · October 2026

Moral Hazard in Multi-Agent Language Models

Dane Malenfant

Question
How do language agents cooperate when gathering useful information is costly, hidden, and primarily benefits another agent?
Contribution
I introduce the Dialogue Moral Hazard Game and use it to evaluate 17 language models. The results show that similar team-level success can emerge from different cooperation mechanisms, motivating evaluations that measure information gathering and disclosure rather than aggregate success alone.
Mean reward comparison for core-triggered, oracle, random-matched, never-switch, and return-trigger control
Preprint Revised v2

arXiv v2 · Revised August 2026 · Presented at World Modeling Workshop 2026

Reinforcing the World's Edge: Continual Learning at the Multi-Agent–World Boundary

Dane Malenfant

Question
How long can success-conditioned reusable structure remain valid when peer learning changes the focal agent's induced rewards and dynamics?
Contribution
I recast decentralized peer learning as an agent-centric continual-RL problem and define an invariant core of high-coverage abstract patterns in successful trajectories. A worst-case-tight stability theorem yields Ω(1/η) survival and conditional Θ(1/η) first-exit laws; registered 64-stream continual-control and cue-MNIST studies show that core erosion predicts failure and enables near-oracle intervention.

Research programme · Ongoing

Credit Assignment & Compositional Learning

How can learning systems isolate the decisions and representations that matter, then reuse them across long horizons or unseen combinations?

Reward curves for ConSpec with PPO and three reinforcement-learning baselines
Published Code available

NeurIPS 2023 · New Orleans, USA · December 2023

Contrastive Retrospection: honing in on critical steps for rapid learning and generalization in RL

Chen Sun, Wannan Yang, Thomas Jiralerspong, Dane Malenfant, Benjamin Alsbury-Nealy, Yoshua Bengio, and Blake Aaron Richards

Question
How can reinforcement-learning agents identify the few temporally distant decisions that determine eventual success?
Contribution
We introduce Contrastive Retrospection (ConSpec), a plug-in method that uses offline contrastive learning to discover interpretable prototypes of critical states and turn them into intrinsic rewards. ConSpec improves learning across diverse tasks and supports out-of-distribution generalization when sensory features change.
Shared encoder-decoder architecture with factor-specific velocity heads
Spotlight Workshop paper

FoGen @ ICML 2026 · Seoul, South Korea · July 2026

Compositional Flow Matching with Factored Velocity Fields

Avery Hee-Woon Ryoo, Dane Malenfant, Matthew G. Perich, and Guillaume Lajoie

Question
How can conditional generative models generalize to novel combinations of factors that were absent from training?
Contribution
We propose a factored conditional flow-matching architecture with a shared base velocity and factor-specific heads. It matches or outperforms a monolithic baseline on zero-shot combinations, substantially lowers held-out FID on the hardest Shapes3D splits, and supports adding a new factor without retraining existing components.

Interdisciplinary programme

AI in Society & Culture

How do technical systems interact with institutions, public policy, language, and the narratives through which people interpret causality?

Stacked policy-strategy coverage bars for eleven NATO member states
Workshop paper Poster available

Harms and Risks of AI in the Military 2024 · Montréal, Canada · December 2024

Inconsistencies in Artificial Intelligence Strategy Alignment of NATO Member States

Itai Epstein, Dane Malenfant, Sara Parker, and Cella Wardrop

Question
How consistently have NATO member states publicly articulated policies for military uses of artificial intelligence?
Contribution
We build a comparative dataset from official national strategies and public statements. Although 88% of NATO members had a national AI strategy, only 34% had published military-specific AI policy, revealing a substantial transparency and alignment gap.
Random Forest feature weights for literary and SemEval causal-relation data
Published

Text2Story @ ECIR 2022 · Stavanger, Norway · April 2022

Causality Mining in Fiction

Margaret Meehan, Dane Malenfant, and Andrew Piper

Question
Do causal-relation models developed on standard NLP benchmarks transfer reliably to literary fiction?
Contribution
We introduce a labeled dataset containing 548 positive causal event pairs from 141 contemporary literary passages, together with matched negative examples. Comparing feature-based and BERT models against SemEval data reveals linguistic and domain differences that complicate causal-relation extraction from fiction.