16 papers
The Usefulness Gap in Proof-of-Useful-Work: An Empirical Study of Pearl's cuPOW Protocol
Abhinaba Basu
Pearl, a Layer-1 blockchain with high-profile AI industry endorsements, markets its Proof-of-Useful-Work (PoUW) protocol as simultaneously securing the network and performing AI in…
HubRouter: A Pluggable Sub-Quadratic Routing Primitive for Hybrid Sequence Models
Abhinaba Basu
We introduce HubRouter, a pluggable module that replaces O(n^2) attention layers with O(nM) hub-mediated routing, where M << n is a small number of learned hub tokens. We demonstra…
When Does Content-Based Routing Work? Representation Requirements for Selective Attention in Hybrid Sequence Models
Abhinaba Basu
We identify a routing paradox in hybrid sequence models: content-based routing - deciding which tokens deserve expensive attention - requires pairwise computation, and this require…
Transactional Attention: Semantic Sponsorship for KV-Cache Retention
Abhinaba Basu
At K=16 tokens (0.4% of a 4K context), every existing KV-cache compression method achieves 0% on credential retrieval. The failure mode is dormant tokens: credentials, API keys, an…
Measuring and curing reasoning rigidity: from decorative chain-of-thought to genuine faithfulness
Abhinaba Basu, Pavan Chakraborty
Language models increasingly show their work by writing step-by-step reasoning before answering. But are these steps genuinely used, or is the answer rigid - fixed before reasoning…
Accelerated Dopant Screening in Oxide Semiconductors via Multi-Fidelity Contextual Bandits and a Three-Tier DFT Validation Funnel
Abhinaba Basu
Band gap engineering of oxide semiconductors through doping is critical for photocatalysis and optoelectronics, yet the combinatorial space of dopant elements, substitution sites,…
The Coordinate System Problem in Persistent Structural Memory for Neural Architectures
Abhinaba Basu
We introduce the Dual-View Pheromone Pathway Network (DPPN), an architecture that routes sparse attention through a persistent pheromone field over latent slot transitions, and use…
Why Agent Caching Fails and How to Fix It: Structured Intent Canonicalization with Few-Shot Learning
Abhinaba Basu
Personal AI agents incur substantial cost via repeated LLM calls. We show existing caching methods fail: GPTCache achieves 37.9% accuracy on real benchmarks; APC achieves 0-12%. Th…
When Names Change Verdicts: Intervention Consistency Reveals Systematic Bias in LLM Decision-Making
Abhinaba Basu, Pavan Chakraborty
Large language models (LLMs) are increasingly used for high-stakes decisions, yet their susceptibility to spurious features remains poorly characterized. We introduce ICE-Guard, a…
Proof-Carrying Materials: Falsifiable Safety Certificates for Machine-Learned Interatomic Potentials
Abhinaba Basu, Pavan Chakraborty
Machine-learned interatomic potentials (MLIPs) are deployed for high-throughput materials screening without formal reliability guarantees. We show that a single MLIP used as a stab…
Budget-Sensitive Discovery Scoring: A Formally Verified Framework for Evaluating AI-Guided Scientific Selection
Abhinaba Basu, Pavan Chakraborty
Scientific discovery increasingly relies on AI systems to select candidates for expensive experimental validation, yet no principled, budget-aware evaluation framework exists for c…
Tool Receipts, Not Zero-Knowledge Proofs: Practical Hallucination Detection for AI Agents
Abhinaba Basu
AI agents that execute tasks via tool calls frequently hallucinate results - fabricating tool executions, misstating output counts, or presenting inferences as facts. Recent approa…
Cross-Domain Uncertainty Quantification for Selective Prediction: A Comprehensive Bound Ablation with Transfer-Informed Betting
Abhinaba Basu
We present a comprehensive ablation of nine finite-sample bound families for selective prediction with risk control, combining concentration inequalities (Hoeffding, Empirical Bern…
Structural Sensitivity in Compressed Transformers: Relative Error Propagation and Layer Removal
Abhinaba Basu, Kumkum Basu, Koushik Deb
Compressing transformer weights makes large language models cheaper to deploy. But each layer's compression introduces an error. These errors accumulate as the signal passes throug…
ICE: Intervention-Consistent Explanation Evaluation with Statistical Grounding for LLMs
Abhinaba Basu, Pavan Chakraborty
Evaluating whether explanations faithfully reflect a model's reasoning remains an open problem. Existing benchmarks use single interventions without statistical testing, making it…
Contextual StereoSet: Stress-Testing Bias Alignment Robustness in Large Language Models
Abhinaba Basu, Pavan Chakraborty
A model that avoids stereotypes in a lab benchmark may not avoid them in deployment. We show that measured bias shifts dramatically when prompts mention different places, times, or…