RH
Rajdeep Haldar
cs.LGstat.MLcs.AIcs.CRcs.CYphysics.comp-phphysics.plasm-ph
On Valency
published · living versionsW_y6sxdjsj·v1 · currentpublished
Preprints & journals
7 papers in the corpus · 2023–2026Divergence based Ranked Preference Alignment for LLMs2610.25588v1 · Rajdeep Haldar, Yue Xing, Qifan Song2026 · 0 citationsarXivon Valency
PHASE: Multi-Regime Modeling of Incompressible Magnetohydrodynamics2609.37609v1 · Radhika Achikanath Chirakkara, Rajdeep Haldar, Zezheng Song et al.2026 · 0 citationsarXiv
f-GRPO and Beyond: Divergence-Based Reinforcement Learning Algorithms for General LLM Alignment2602.05946v3 · Rajdeep Haldar, Lantao Mei, Guang Lin et al.2026 · 0 citationsarXiv
LLM Safety Alignment is Divergence Estimation in Disguise2502.00657v3 · Rajdeep Haldar, Ziyi Wang, Qifan Song et al.2025 · 0 citationsarXiv
Adversarial Vulnerability as a Consequence of On-Manifold Inseparibility2410.06921v1 · Rajdeep Haldar, Yue Xing, Qifan Song et al.2024 · 0 citationsarXiv
Effect of Ambient-Intrinsic Dimension Gap on Adversarial Vulnerability2403.03967v2 · Rajdeep Haldar, Yue Xing, Qifan Song2024 · 0 citationsarXiv
On Neural Network approximation of ideal adversarial attack and convergence of adversarial training2307.16099v1 · Rajdeep Haldar, Qifan Song2023 · 1 citationarXiv
Career total: 10 works. 7 are in this corpus.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.