Dimensions overall score 7.0

GitHub Code Pulse

No public code linked for this paper yet.

• Heddle achieves up to 2.5x higher end-to-end rollout throughput compared to state-of-the-art baselines.result 90%
• Rollout generation is the dominant bottleneck in agentic RL, consuming over 80% of the entire training time.technical 90%
• Frequent tool calls in agentic RL induce a long-tailed distribution of trajectory generation times, creating a straggler effect.technical 85%
• Heddle's trajectory-level scheduling uses runtime prediction and progressive priority to minimize cumulative queueing delays.method 85%
• Heddle employs trajectory-aware placement via presorted dynamic programming and opportunistic migration to minimize interference overhead.method 85%
• Heddle's trajectory-adaptive resource manager dynamically tunes model parallelism to accelerate per-token time for long-tail trajectories while maintaining high throughput for short ones.method 85%
• Heddle masks migration overhead by transmitting data asynchronously during tool-call intervals, keeping the critical execution path unblocked.method 80%
• In agentic RL, high sampling temperatures are often employed to encourage exploration, inherently inducing high output variance in trajectory length.technical 80%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card