8 papers
Swiss-Knife: A Framework for Reconfigurable Externalised Multi-Objective Alignment at Decode Time
Agnibh Karmakar, Mayur Parvatikar, Shreyash Dhoot +5
Decode-time alignment methods steer a frozen language model by scoring candidate continuations with an external reward and selecting the maximiser. We argue that this shared design…
MENTIS: What Belief Changes Under Alignment? Measuring Multi-Scale Latent Torsion in Language Models
Partha Pratim Saha, Samarth Raina, Mayur Parvatikar +4
Preference alignment has substantially improved the observable behavior of large language models, yet it remains unclear what alignment changes internally. Aligned systems still fa…
SPHERICAL KV: Angle-Domain Attention and Rate-Distortion Retention for Efficient Long-Context Inference
Anay Chauhan, Gurucharan Marthi Krishna Kumar, Arion Das +4
Long-context inference is increasingly constrained by the KV cache: resident memory grows with context length, and decoding becomes limited by repeated High Bandwidth Memory (HBM)…
Safe Equilibrium Policy Optimization for Strategic Agent Policies
Karthika Arumugam, Kiran Kumar Manku, Amit Dhanda
Language models fine-tuned with reinforcement learning typically optimize for task reward, ignoring multi-agent strategic structure. Because these agents condition on natural langu…
Chart-RL: Policy Optimization Reinforcement Learning for Enhanced Visual Reasoning in Chart Question Answering with Vision Language Models
Yunfei Bai, Amit Dhanda, Shekhar Jain
The recent advancements in Vision Language Models (VLMs) have demonstrated progress toward true intelligence requiring robust reasoning capabilities. Beyond pattern recognition, li…
DeltaLogic: Minimal Premise Edits Reveal Belief-Revision Failures in Logical Reasoning Models
Amit Dhanda
Reasoning benchmarks typically evaluate whether a model derives the correct answer from a fixed premise set, but they under-measure a closely related capability that matters in dyn…
SPINAL -- Scaling-law and Preference Integration in Neural Alignment Layers
Arion Das, Partha Pratim Saha, Amit Dhanda +3
Direct Preference Optimization (DPO) is a principled, scalable alternative to RLHF for aligning large language models from pairwise preferences, but its internal geometric footprin…
Multi-Dimensional Summarization Agents with Context-Aware Reasoning over Enterprise Tables
Amit Dhanda
We propose a novel framework for summarizing structured enterprise data across multiple dimensions using large language model (LLM)-based agents. Traditional table-to-text models o…