1 citations · 1 across the 6 of their papers we have counts for
9 papers
Projected Distillation: On Learning to Solve Hard Problems from Close, Correct Targets
Shripad V. Deshmukh, Christian Bock, Kangrui Ruan +4
On-policy learning is central to post-training: training on a model's own generations limits distribution shift and helps preserve existing capabilities. On hard problems, however,…
Reward-Aligned Distillation
Yilan Chen, Zian Su, Manish Prajapat +4
Reinforcement learning with verifiable rewards improves the reasoning capabilities of large language models, but remains computationally expensive and sample inefficient. On-policy…
Token Utility for On-Policy Distillation via Local Trust-Region Optimization
Narayanan Sadagopan, Yilan Chen, Vijay Lingam
Token-selection methods for on-policy distillation (OPD) prioritize student-visited positions using diverse signals, including uncertainty, teacher-student disagreement, and mismat…
AHEAD: Adaptive Hindsight with Environment-Augmented Distillation for Agentic RL
Xiaolong Jin, Dingmin Wang, Vijay Lingam +1
Training multi-turn LLM agents with reinforcement learning typically relies on trajectory-level rewards, which assign a uniform advantage to every step and cannot identify which de…
Reward-Gated On-Policy Distillation
Mohammad Sadegh Akhondzadeh, Vijay Lingam, Atula Tejaswi +3
On-policy distillation is a powerful way to transfer reasoning ability from a strong teacher to a smaller student: the student samples trajectories from its own policy, and the tea…
SCATR: Simple Calibrated Test-Time Ranking
Divya Shyamal, Marta Knežević, Lan Tran +3
Test-time scaling (TTS) improves large language models (LLMs) by allocating additional compute at inference time. In practice, TTS is often achieved through parallel scaling: gener…
ExecTune: Effective Steering of Black-Box LLMs with Guide Models
Vijay Lingam, Aditya Golatkar, Anwesan Pal +6
For large language models deployed through black-box APIs, recurring inference costs often exceed one-time training costs. This motivates composed agentic systems that amortize exp…
MURPHY: Feedback-Aware GRPO with Retrospective Credit Assignment for Multi-Turn Code Generation
Chanakya Ekbote, Vijay Lingam, Sujay Sanghavi +4
Reinforcement Learning with Verifiable Rewards (RLVR) has become a standard recipe for post-training LLMs on reasoning tasks, with Group Relative Policy Optimization (GRPO) emergin…
SVFT: Parameter-Efficient Fine-Tuning with Singular Vectors
Vijay Lingam, Atula Tejaswi, Aditya Vavre +7
Popular parameter-efficient fine-tuning (PEFT) methods, such as LoRA and its variants, freeze pre-trained model weights \(W\) and inject learnable matrices \(ΔW\). These \(ΔW\) mat…