archive
Every paper Pith has read. Search by title, abstract, or pith.
527 papers in cs.GT · page 3
-
Algorithm regret scales with corruption in bilateral trade
Regret Minimization in Bilateral Trade With Perturbed Markets
-
Team-symmetric games always admit a symmetric Nash equilibrium
DelAC: A Multi-agent Reinforcement Learning of Team-Symmetric Stochastic Games
-
(k+1)/(k+2)-EFkX allocations exist for any agents when k>2
Approximate Envy-Free Allocations up to any $k$ Goods
-
Vote-Left triples Faithful win rate in The Traitors
The Vote-Left Equilibrium: A Deterministic Coordination Strategy for the Faithful in The Traitors
-
Model-free RL learns near-SNE policies in Karma economies
Towards Model-Free Learning in Dynamic Population Games: An Application to Karma Economies
-
Ranking certificates prove wins in infinite polynomial games
Automated Approach for Solving Infinite-state Polynomial Reachability Games
-
Resource allocation equilibria are identity or alternating flat functions
A Resource Allocation Game and its Equilibrium Strategies
-
The paper introduces risk-sensitive games for preference learning in LLMs
Structure from Strategic Interaction & Uncertainty: Risk Sensitive Games for Robust Preference Learning
-
Risk-sensitive preference games preserve monotonicity via translation-invariant risk…
Structure from Strategic Interaction & Uncertainty: Risk Sensitive Games for Robust Preference Learning
-
Maximal EF1 always exists for two monotone agents on any graph
Fair Allocation under Conflict Constraints
-
Every vitality index extends uniquely to node groups via group Shapley value
Group Vitality Indices: Axioms and Algorithms
-
Backward LPs yield optimal recommendations to strategic agents
Action Recommendations for Sequentially Rational Strategic Agents
-
Prediction markets adapt liquidity using online learning
Adaptive Liquidity in Prediction Markets via Online Learning
-
Greedy cuts queries for ensemble selection with 1-1/e guarantee
Efficient Ensemble Selection from Binary and Pairwise Feedback
-
Polynomial-time method finds exact SPPEs for constant goods
Pacing Equilibria in Second-Price Auctions with Few Goods
-
Weighted graphs get nearly equitable colorings with O(Δ) colors
Equitable Colorings of Vertex-Weighted Graphs
-
Allocations beat 1/4 MMS bound for large agent counts
On MMS, APS and XOS
-
Willed agents catalyze cooperation where rational maximizers fail
Modeling Decision-Making with Will for Cooperation in Social Dilemmas
-
Separate LoRA adapters restore adversarial pressure in safety self-play
The Attacker in the Mirror: Breaking Self-Consistency in Safety via Anchored Bipolicy Self-Play
-
Prosocial agents recover welfare losses no mechanism can fix
Mechanism Design Is Not Enough: Prosocial Agents for Cooperative AI
-
Longer memory reduces cooperation among LLM agents
The Memory Curse: How Expanded Recall Erodes Cooperative Intent in LLM Agents
-
Nash equilibria defined from ordinal rankings alone
Nash without Numbers: A Social Choice Approach to Mixed Equilibria in Context-Ordinal Games
-
Zero-determinant strategies match SSE defense performance
Zero-determinant Strategy for Moving Target Defense: Existence, Performance, and Computation
-
Strategic responses make uniform privacy budgets suboptimal
Differentially Private Auditing Under Strategic Response
-
Non-affine approvals force miscalibration in all proper scoring rules
The Endogeneity of Miscalibration: Impossibility and Escape in Scored Reporting
-
Quotient semivalues block false-name attacks on data attribution
Quotient Semivalues for False-Name-Resistant Data Attribution
-
Cost reports plus assignments stop wasted LLM data subsidies
Incentivizing User Data Contributions for LLM Improvement under Withdrawal Rights
-
Fee sharing equilibria beat random allocation on DAG throughput
Game-Theoretic Analysis of Transaction Selection in DAG-Based Distributed Ledgers
-
Game theory balances conflicting LLM goals at inference time
Common-agency Games for Multi-Objective Test-Time Alignment
-
Valuation design cuts Tullock optimization to two variables
Incentive Design in Competitive Resource Allocation: Exploiting Valuation Asymmetry in Tullock Contests
-
Response times identify average preferences from single anonymous choices
Response Time Enhances Alignment with Heterogeneous Preferences
-
Least core cuts LLM calls for creator credit assignment
In-Context Credit Assignment via the Core
-
Auctions embed ads in LLMs without distorting content
Mechanism Design for Quality-Preserving LLM Advertising
-
Adaptive scalarization choice steers vector games toward preferred equilibria
Online Scalarization in Vector-Valued Games
-
Online scalarization choice yields sublinear regret in vector-valued games
Online Scalarization in Vector-Valued Games
-
LLMs crowd creative ideas below human parity
Ex Ante Evaluation of AI-Induced Idea Diversity Collapse
-
Folk theorem for LLMs sustains any rational outcome as ε-equilibrium
Sustaining Cooperation in Populations Guided by AI: A Folk Theorem for LLMs
-
Sequential trials with subsidies raise social utility over 35%
Optimizing Social Utility in Sequential Experiments
-
EFX fails even for symmetric submodular valuations with three agents
Counterexamples to EFX for Submodular and Subadditive Valuations
-
Agents learn Nash equilibria independently with local observations
Independent Learning of Nash Equilibria in Partially Observable Markov Potential Games with Decoupled Dynamics
-
Static audit policies leave unclosable gaps for strategic gaming
A Benchmark for Strategic Auditee Gaming Under Continuous Compliance Monitoring
-
Core committees exist for approval elections with up to five voters
Core Existence in Approval-Based Committee Elections with up to Five Voter Types
-
Feedback on summaries trains online system to personalize reviews
PREFER: Personalized Review Summarization with Online Preference Learning
-
Method yields undominated ex ante stable school lotteries
A Simple Method for School Choice Lotteries
-
Blockchain procurement loss scales logarithmically with fault costs
Adversarial procurement in blockchains
-
Counterfactual utilities satisfy vNM axioms on potential outcomes
An Axiomatic Foundation for Decisions with Counterfactual Utility
-
Platforms balance profit and welfare using pricing
Pricing, Matching, and Bundling: an Equilibrium Analysis of Online Platforms
-
KFCA rewards federated clients truthfully without ground truth
Knowledge-Free Correlated Agreement for Incentivizing Federated Learning
-
Recognizing graph supports for preferences is NP-hard with few edges
When Graph Traversal Meets Structured Preferences: Unified Framework and Complexity Results
-
PJR+ is the minimal proportionality notion under four mild axioms
An Axiomatic Analysis of Proportionality Notions in Approval-Based Multiwinner Voting