Publications

Supp Fig6+ Legend from PGV001, a Multi-Peptide Personalized Neoantigen Vaccine Platform: Phase I Study in Patients with Solid and Hematologic Malignancies in the Adjuvant Setting
Mansi Saxena
Thomas U. Marron
Julia Kodysh
John P. Finnigan
Sayali Onkar
Anna Kaminska
Kevin Tuballes
Ruiwei Guo
Rachel Lubong Sabado
Marcia Meseck
Timothy J. O’Donnell
Robert Sebra
Samir Parekh
Matthew D. Galsky
Ana Blasquez
Gustavo Gimenez
Mesude Bicak
Cansu Cimen Bozkus
Daniela Delbeau-Zagelbaum
Denise Rodriguez … (see 17 more)
Ana Acuña-Villaorduña
Krzysztof Misiukiewicz
Marshall R. Posner
Brett A. Miles
Hanna Y. Irie
Amy Tiersten
Deborah B. Doroshow
Andrea Wolf
John Mandeli
Rachel Brody
Andres Μ. Salazar
Sacha Gnjatic
Jeff Hammerbacher
Eric E. Schadt
Philip Friedlander
Alex Rubinsteyn
Nina Bhardwaj
<p>Supplementary Fig6. Gating strategy for phenotyping of immune cells by flow cytomtery. EM: Effector memory, TEMRA: Effector memory … (see more)Re-expression RA, CM: Central Memory</p>
2026-06-16
American Association for Cancer Research (accepted)
doi.org
Supp Fig7+ Legend from PGV001, a Multi-Peptide Personalized Neoantigen Vaccine Platform: Phase I Study in Patients with Solid and Hematologic Malignancies in the Adjuvant Setting
Mansi Saxena
Thomas U. Marron
Julia Kodysh
John P. Finnigan
Sayali Onkar
Anna Kaminska
Kevin Tuballes
Ruiwei Guo
Rachel Lubong Sabado
Marcia Meseck
Timothy J. O’Donnell
Robert Sebra
Samir Parekh
Matthew D. Galsky
Ana Blasquez
Gustavo Gimenez
Mesude Bicak
Cansu Cimen Bozkus
Daniela Delbeau-Zagelbaum
Denise Rodriguez … (see 17 more)
Ana Acuña-Villaorduña
Krzysztof Misiukiewicz
Marshall R. Posner
Brett A. Miles
Hanna Y. Irie
Amy Tiersten
Deborah B. Doroshow
Andrea Wolf
John Mandeli
Rachel Brody
Andres Μ. Salazar
Sacha Gnjatic
Jeff Hammerbacher
Eric E. Schadt
Philip Friedlander
Alex Rubinsteyn
Nina Bhardwaj
<p>Supplementary Fig7. Characterizing peripheral cellular immune environment in patients receiving PGV001. Multicolor flowcytometry pe… (see more)rformed to phenotype circulating immune cells in patients and healthy donors (HD). PID-017 was excluded from analysis. a) Tabulated view of age and sex of the seven healthy donors whose PBMCs are used in the study. b) Frequency of lymphoid immune cell subsets over the course of treatment. Each dot represents a subject. # p value indicates un-paired two-sided Student’s T-test comparing HD with patient cohort # <0.05, ## <0.01. c) Frequency of myeloid immune cell subsets over the course of treatment. Each dot represents a subject. d) Depicting Fold change from “Pre” in listed immune cell subsets over the course of treatment. e) Pie charts depicting CD4+ and CD8+ T cell states in patient blood and healthy donors. f) �4+ T cells expressing TIGIT and CTLA4 shown as a fold change from baseline in patient blood. *p value indicates paired two-sided Student’s T-test comparing post treatment samples with “pre”. * <0.05. Data in pie charts depicts median. Patient samples, N=12. Healthy donor samples, N=7. Data in graphs shown as mean with bar graphs showing +/- SEM.</p>
2026-06-16
American Association for Cancer Research (accepted)
doi.org
Supplementary Table1. from PGV001, a Multi-Peptide Personalized Neoantigen Vaccine Platform: Phase I Study in Patients with Solid and Hematologic Malignancies in the Adjuvant Setting
Mansi Saxena
Thomas U. Marron
Julia Kodysh
John P. Finnigan
Sayali Onkar
Anna Kaminska
Kevin Tuballes
Ruiwei Guo
Rachel Lubong Sabado
Marcia Meseck
Timothy J. O’Donnell
Robert Sebra
Samir Parekh
Matthew D. Galsky
Ana Blasquez
Gustavo Gimenez
Mesude Bicak
Cansu Cimen Bozkus
Daniela Delbeau-Zagelbaum
Denise Rodriguez … (see 17 more)
Ana Acuña-Villaorduña
Krzysztof Misiukiewicz
Marshall R. Posner
Brett A. Miles
Hanna Y. Irie
Amy Tiersten
Deborah B. Doroshow
Andrea Wolf
John Mandeli
Rachel Brody
Andres Μ. Salazar
Sacha Gnjatic
Jeff Hammerbacher
Eric E. Schadt
Philip Friedlander
Alex Rubinsteyn
Nina Bhardwaj
<p>Supplementary Table1. Staging at time of enrollment, adjuvant treatment following curative intent treatment until the end of vaccin… (see more)ation and vaccination timing</p>
2026-06-16
American Association for Cancer Research (accepted)
doi.org
Supplementary Table2 from PGV001, a Multi-Peptide Personalized Neoantigen Vaccine Platform: Phase I Study in Patients with Solid and Hematologic Malignancies in the Adjuvant Setting
Mansi Saxena
Thomas U. Marron
Julia Kodysh
John P. Finnigan
Sayali Onkar
Anna Kaminska
Kevin Tuballes
Ruiwei Guo
Rachel Lubong Sabado
Marcia Meseck
Timothy J. O’Donnell
Robert Sebra
Samir Parekh
Matthew D. Galsky
Ana Blasquez
Gustavo Gimenez
Mesude Bicak
Cansu Cimen Bozkus
Daniela Delbeau-Zagelbaum
Denise Rodriguez … (see 17 more)
Ana Acuña-Villaorduña
Krzysztof Misiukiewicz
Marshall R. Posner
Brett A. Miles
Hanna Y. Irie
Amy Tiersten
Deborah B. Doroshow
Andrea Wolf
John Mandeli
Rachel Brody
Andres Μ. Salazar
Sacha Gnjatic
Jeff Hammerbacher
Eric E. Schadt
Philip Friedlander
Alex Rubinsteyn
Nina Bhardwaj
<p>Supplementary Table2 – Variant count per patient</p>
2026-06-16
American Association for Cancer Research (accepted)
doi.org
Supplementary Table3 from PGV001, a Multi-Peptide Personalized Neoantigen Vaccine Platform: Phase I Study in Patients with Solid and Hematologic Malignancies in the Adjuvant Setting
Mansi Saxena
Thomas U. Marron
Julia Kodysh
John P. Finnigan
Sayali Onkar
Anna Kaminska
Kevin Tuballes
Ruiwei Guo
Rachel Lubong Sabado
Marcia Meseck
Timothy J. O’Donnell
Robert Sebra
Samir Parekh
Matthew D. Galsky
Ana Blasquez
Gustavo Gimenez
Mesude Bicak
Cansu Cimen Bozkus
Daniela Delbeau-Zagelbaum
Denise Rodriguez … (see 17 more)
Ana Acuña-Villaorduña
Krzysztof Misiukiewicz
Marshall R. Posner
Brett A. Miles
Hanna Y. Irie
Amy Tiersten
Deborah B. Doroshow
Andrea Wolf
John Mandeli
Rachel Brody
Andres Μ. Salazar
Sacha Gnjatic
Jeff Hammerbacher
Eric E. Schadt
Philip Friedlander
Alex Rubinsteyn
Nina Bhardwaj
<p>Supplementary Table3: List of mutated peptide IDs and peptide sequence in each patients’ vaccine</p>
2026-06-16
American Association for Cancer Research (accepted)
doi.org
Supplementary Table4 from PGV001, a Multi-Peptide Personalized Neoantigen Vaccine Platform: Phase I Study in Patients with Solid and Hematologic Malignancies in the Adjuvant Setting
Mansi Saxena
Thomas U. Marron
Julia Kodysh
John P. Finnigan
Sayali Onkar
Anna Kaminska
Kevin Tuballes
Ruiwei Guo
Rachel Lubong Sabado
Marcia Meseck
Timothy J. O’Donnell
Robert Sebra
Samir Parekh
Matthew D. Galsky
Ana Blasquez
Gustavo Gimenez
Mesude Bicak
Cansu Cimen Bozkus
Daniela Delbeau-Zagelbaum
Denise Rodriguez … (see 17 more)
Ana Acuña-Villaorduña
Krzysztof Misiukiewicz
Marshall R. Posner
Brett A. Miles
Hanna Y. Irie
Amy Tiersten
Deborah B. Doroshow
Andrea Wolf
John Mandeli
Rachel Brody
Andres Μ. Salazar
Sacha Gnjatic
Jeff Hammerbacher
Eric E. Schadt
Philip Friedlander
Alex Rubinsteyn
Nina Bhardwaj
<p>Supplementary Table4: Number of Subjects who had Treatment Emergent Adverse Events by grade, system (13 subjects)</p>
2026-06-16
American Association for Cancer Research (accepted)
doi.org
Supplementary Table5 from PGV001, a Multi-Peptide Personalized Neoantigen Vaccine Platform: Phase I Study in Patients with Solid and Hematologic Malignancies in the Adjuvant Setting
Mansi Saxena
Thomas U. Marron
Julia Kodysh
John P. Finnigan
Sayali Onkar
Anna Kaminska
Kevin Tuballes
Ruiwei Guo
Rachel Lubong Sabado
Marcia Meseck
Timothy J. O’Donnell
Robert Sebra
Samir Parekh
Matthew D. Galsky
Ana Blasquez
Gustavo Gimenez
Mesude Bicak
Cansu Cimen Bozkus
Daniela Delbeau-Zagelbaum
Denise Rodriguez … (see 17 more)
Ana Acuña-Villaorduña
Krzysztof Misiukiewicz
Marshall R. Posner
Brett A. Miles
Hanna Y. Irie
Amy Tiersten
Deborah B. Doroshow
Andrea Wolf
John Mandeli
Rachel Brody
Andres Μ. Salazar
Sacha Gnjatic
Jeff Hammerbacher
Eric E. Schadt
Philip Friedlander
Alex Rubinsteyn
Nina Bhardwaj
<p>Supplementary Table5: Cellular Immunophenotyping antibody panels</p>
2026-06-16
American Association for Cancer Research (accepted)
doi.org
Adaptive Multi-Horizon Reinforcement Learning
Manoosh Samiei
Doina Precup
Paul Masset
Effective decision-making in complex and changing environments requires balancing short-term and long-term consequences. In reinforcement le… (see more)arning (RL), this trade-off is typically controlled through a fixed discount factor, which imposes a single exponentially discounted temporal horizon. However, biological agents exhibit flexible and adaptive temporal discounting, suggesting that effective planning requires multiple timescales. Here, we propose a multi-horizon approach that adaptively selects and combines temporal horizons, enabling robust adaptation to changes in reward structure without manual discount-factor tuning. This flexibility makes the method particularly suitable for continual learning scenarios involving task switches and varying environmental configurations. Empirically, we demonstrate that our approach identifies effective discount factors across a range of MiniGrid environments, including continual settings composed of three sequentially changing tasks. These results suggest that adaptive temporal discounting can improve parameter efficiency and enhance adaptability in both artificial and biologically inspired learning systems.
2026-06-15
ContinualRL @ Reinforcement Learning Conference (published)
doi.org
openreview.net
Bayesian Decision-Time Inference for In-Context Reinforcement Learning from Suboptimal Data
Anaïs Berkes
Vincent Taboga
Donna Vakalis
David Rolnick
Yoshua Bengio
In-context reinforcement learning (ICRL) promises rapid adaptation without parameter updates, but standard supervised objectives often fail … (see more)when pretraining data is generated by suboptimal behaviour policies. In these regimes, logged actions are unreliable labels while rewards still provide value-relevant information. To address this, we introduce SPICE, a Bayesian decision-time inference method that shifts online ICRL from action-logit prediction to approximate posterior inference over action values, requiring neither expert action labels nor algorithmic learning traces. SPICE learns a task-conditioned value prior with a transformer value ensemble and, at test time with parameters frozen, fuses this prior with kernel-weighted context evidence via a closed-form Gaussian fusion update. The resulting estimates drive a posterior-UCB controller, enabling principled online exploration and adaptation without gradient updates. A stochastic-bandit analysis shows logarithmic regret growth for the fixed-prior controller under scheduled exploration, while quantifying the additional early cost caused by inaccurate prior estimates. Across bandits, Darkroom, image-based MiniWorld, and continuous building control, SPICE adapts more effectively from suboptimal data than supervised ICRL baselines.
2026-06-15
ContinualRL @ Reinforcement Learning Conference (published)
openreview.net
openreview.net
A Bellman Optimality Equation for Plasticity
Jeremy Lucas
Doina Precup
In continual reinforcement learning, carefully managing the stability-plasticity tradeoff remains a core challenge. Recent work by Abel et a… (see more)l. (2025) formalized this dilemma by defining plasticity as the generalized directed information from an agent’s observations to its actions, and empowerment as the generalized directed information from its actions to its observations. This formulation successfully reframes the traditional stability-plasticity tradeoff as an empowerment-plasticity tradeoff. However, while extensive literature exists on optimizing for empowerment, there is currently no research addressing the optimization of plasticity under this new definition. This paper presents preliminary work toward optimizing plasticity within Markov decision processes. We show that there exists a Bellman optimality equation for optimizing plasticity similar to previous work for empowerment.
2026-06-15
ContinualRL @ Reinforcement Learning Conference (published)
openreview.net
openreview.net
Beyond the GUI Paradigm: Do Mobile Agents Need the Phone Screen?
Li Gu
Zihuan Jiang
Linqiang Guo
Zhixiang Chi
Ziqiang Wang
Huan Liu
Yuanhao Yu
Tse-Hsun Chen
Yang Wang
Recent advances in mobile agents are dominated by the GUI paradigm, in which agents perceive UI information and emit screen interactions. Ho… (see more)wever, mobile platforms also expose a command-line interface (CLI) that provides direct access to device services and data. We argue CLI deserves first-class consideration alongside GUI. We evaluate three coding agents (Claude Code, Terminus-2, mini-swe-agent) across four model APIs on AndroidWorld and MobileWorld without any mobile-specific post-training, comparing against three reproducible GUI baselines (GUI-Owl-1.5-32B, MAI-UI, Qwen3-VL-32B). Claude Code (Opus 4.7) reaches 71.8\% and 51.9\%, outperforming every reproducible GUI baseline (69.3/68.1/57.8\% on AndroidWorld; 43.2/26.3/13.3\% on MobileWorld), while every other CLI configuration remains competitive. To establish the paradigm's ceiling, we provide oracle CLI solutions that reach 88.8\% on AndroidWorld (103/116 tasks CLI-solvable) and 86.3\% on MobileWorld (101/117 tasks CLI-solvable), indicating substantial room for future improvement. To cover everyday user intents beyond the GUI scope, we introduce the \textbf{CLI-Advantage Task Suite}, comprising 45 templates across five categories: bulk operations, multi-condition filtering, aggregation, cross-app workflows, and hidden device state. Every CLI agent outperforms every GUI baseline in all five categories, with substantially fewer steps per task (10.7 vs.\ 18.6). To support future research on mobile CLI agents, we will open-source agent implementations, oracle solutions, the CLI-Advantage suite, and evaluation infrastructure.
2026-06-15
arXiv (preprint)
doi.org
arxiv.org
Detecting Naturally Emerging Temporal Abstractions
Jayesh Khullar
Francois Rivest
Breaking down large tasks into smaller sub-tasks, either to accelerate learning or enable transfer across related environments, remains a ce… (see more)ntral challenge in reinforcement learning (RL). One way Hierarchical Reinforcement Learning addresses this is by introducing temporal abstractions, often instantiated as options: temporally extended action sequences directed toward sub-goals. While prior work largely focuses on algorithms that explicitly learn such options, we ask a different question: can temporal abstractions emerge naturally within general deep reinforcement learning agents? To this end, we introduce Decorrelate Cluster Temporal Activation (DCTA) Analysis, a tool for detecting temporal abstractions in agents that do not explicitly model them. We validate this approach on a custom Four-Room environment and Atari benchmarks. Most importantly, we show that DQN and PPO naturally develop internal representations with semi-Markov consistent temporal structure — the defining statistical property of temporal abstractions — without explicit option learning objective.
2026-06-15
ContinualRL @ Reinforcement Learning Conference (published)
openreview.net
openreview.net