We have 11 of 66 papers
We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.
Not the right person? Other researchers publish under this name.
Modern LLM post-training composes supervised fine-tuning (SFT), reinforcement learning with verifiable rewards (RLVR), and on-policy distillation (OPD) into multi-stage pipelines, yet these stages are typically designed and evaluated in isolation. We show that this composition is consequential: a stage that improves th...
Just-in-Time Memory (JitMem) consistently outperforms no-memory agents as well as heuristic and learned write-time memory methods, improving over the strongest baseline by 16.2, 16.3, and 3.9 absolute success-rate points, respectively.
Four parallelism plans are left unbounded by the parallelism plans in common use, and each grows differently: expert dispatch with the routing matrix, the vocabulary projection with tokens times vocabulary, gradient checkpoint boundaries with depth times sequence length, and optimizer state with parameter count.
Modern LLM-based agents operate through a harness of tools, reusable skills, and specialist agents that shapes what they observe and what they can do. In practice, this harness continually evolves as new capabilities are added. We introduce EvoHarnessBench, a benchmark for evaluating agents under controlled harness evo...
Experiments spanning mathematical reasoning, multi-domain STEM, code generation, and multi-turn agentic tasks show that RISE outperforms RLVR-only training and on-policy self-distillation across all settings.
This work introduces VisEditBench, a benchmark of 1,395 human-annotated visualization code-editing tasks grounded in realistic visualization workflows and failure cases, and proposes VisEditAgent, a render-grounded editing framework that iteratively generates, executes, validates, and refines candidate edits.
It is found that process-level verification does not consistently improve performance and frequently exhibits high variance, highlighting the difficulty of reliably evaluating partial multi-agent trajectories.
The results suggest that dense credit assignment through distillation can be effective when its likelihood-based scores are empirically validated as meaningful proxies for outcome-relevant credit, when this alignment does not hold.
DSAgentBench is introduced, the first benchmark to evaluate whether agents can automate full data-science workflows inside real computer environments, and reveals a substantial capability gap between current agentic systems and real data-science workflows.
River, a simple training recipe that improves reward quality by filtering low-quality environments and augmenting outcome rewards with process-level behavior regularization is proposed, which achieves the best performance among evaluated open-source RL-trained 8B models across four terminal-agent benchmarks.
Procedural Memory Distillation is proposed, which converts crossepisode signals into reusable procedural memory and distills it into the policy's weights during training, yielding a memory-free model at inference.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
