feat: support GDPO (New)

examples/configs/gdpo_math_1B.yaml

@@ -0,0 +1,62 @@
+# GDPO: inherits from grpo_math_1B.yaml and overrides only what differs.
+defaults: grpo_math_1B.yaml
+
+grpo:
+  adv_estimator:
+    name: "gdpo"
+    normalize_rewards: true
+    use_leave_one_out_baseline: false
+
+checkpointing:
+  checkpoint_dir: "results/gdpo"
+
+policy:
+  model_name: "Qwen/Qwen2.5-1.5B-Instruct"
+  logprob_batch_size: 4
+  max_total_sequence_length: 1024
+  megatron_cfg:
+    optimizer:
+      weight_decay: 0.0
+    scheduler:
+      lr_decay_style: "cosine"
+      lr_warmup_iters: 10
+
+# GDPO uses a single flat data config (GSM8K + math_gdpo_data_processor); replace parent's train/validation/default.
+data:
+  _override_: true
+
+  max_input_seq_length: ${policy.max_total_sequence_length}
+  shuffle: true
+  num_workers: 1
+
+  use_multiple_dataloader: false
+
+  train:
+    dataset_name: "gsm8k"
+    split: train
+  validation:
+    dataset_name: "gsm8k"
+    split: test
+
+  default:
+    prompt_file: null
+    system_prompt_file: "examples/prompts/gsm8k.txt"
+    processor: "math_gdpo_data_processor"
+    env_name: "math_multi_reward"
+
+env:
+  math_multi_reward:
+    num_workers: 8
+    math_verify_impl: "hf_math_verify"
+
+logger:
+  wandb_enabled: true
+  wandb:
+    project: "gdpo-dev"
+    name: "gdpo-dev-logger"
+  swanlab:
+    project: "gdpo-dev"
+    name: "gdpo-dev-logger"
+  mlflow:
+    experiment_name: "gdpo-dev"
+    run_name: "gdpo-dev-logger"

examples/prompts/gsm8k.txt

@@ -0,0 +1,17 @@
+You are a helpful AI assistant.
+
+For every request, you should carefully think through the math problem step by step, then provide the final answer in integer format.
+
+Steps for Each Request:
+1. Think: Provide detailed, step-by-step reasoning, calculations, or derivations.
+2. Produce Final Answer: After step-by-step reasoning, output the final answer in integer format.
+
+Output Format:
+<think>Your thoughts and reasoning</think>
+<answer>Final answer in integer format</answer>
+
+Important Notes:
+1. You must include your reasoning steps inside <think>.
+2. You must always output the Final Answer within <answer> after the reasoning steps is done.
+3. You should consistently work through the solution step by step before giving the final answer.
+4. The final answer can only be an integer.

examples/run_grpo.py

@@ -139,6 +139,13 @@ def main() -> None:
                 "use_multiple_dataloader is not supported with async GRPO"
             )
 
+        # Async GDPO is not supported
+        if config["grpo"]["adv_estimator"]["name"] == "gdpo":
+            raise NotImplementedError(
+                "GDPO is not supported for async training, "
+                "please set grpo.async_grpo.enabled to false in your config."
+            )
+
         from nemo_rl.algorithms.grpo import async_grpo_train
 
         print("🚀 Running async GRPO training")

nemo_rl/algorithms/advantage_estimator.py

@@ -16,6 +16,7 @@
 
 This module provides different advantage estimation strategies:
 - GRPOAdvantageEstimator: Standard GRPO advantage with leave-one-out baseline
+- GDPOAdvantageEstimator: Multi-reward GDPO (per-component baselines, sum then normalize)
 - ReinforcePlusPlusAdvantageEstimator: Reinforce++ with optional baseline subtraction (minus_baseline) and KL penalty in reward
 Reference papers:
 - ProRLv2: https://developer.nvidia.com/blog/scaling-llm-reinforcement-learning-with-prolonged-training-using-prorl-v2/
@@ -24,7 +25,11 @@
 
 import torch
 
-from nemo_rl.algorithms.utils import calculate_baseline_and_std_per_prompt, calculate_kl
+from nemo_rl.algorithms.utils import (
+    calculate_baseline_and_std_per_prompt,
+    calculate_kl,
+    get_gdpo_reward_component_keys,
+)
 
 
 class GRPOAdvantageEstimator:
@@ -37,14 +42,21 @@ def __init__(self, estimator_config: dict, loss_config: dict):
         self.use_leave_one_out_baseline = estimator_config["use_leave_one_out_baseline"]
         self.normalize_rewards = estimator_config["normalize_rewards"]
 
-    def compute_advantage(self, prompt_ids, rewards, mask, **kwargs):
+    def compute_advantage(
+        self,
+        prompt_ids,
+        rewards,
+        repeated_batch,
+        mask,
+        **kwargs,
+    ):
         """Compute GRPO advantages.
 
         Args:
             prompt_ids: Tensor of shape [batch_size] identifying which prompt each sample belongs to.
             rewards: Tensor of shape [batch_size] containing reward for each sample.
+            repeated_batch: Batch (unused; for interface consistency).
             mask: Response token mask of shape [batch_size, seq_len], 1 for valid response tokens, 0 for padding.
-                  Used only for expanding advantages to token-level shape.
             **kwargs: Additional arguments (unused).
 
         Returns:
@@ -69,6 +81,79 @@ def compute_advantage(self, prompt_ids, rewards, mask, **kwargs):
         return advantages.expand(mask.shape)
 
 
+class GDPOAdvantageEstimator:
+    """GDPO-style advantage estimator with leave-one-out baseline.
+
+    Note: GDPO computes advantages for each reward separately over all responses for each prompt.
+    """
+
+    def __init__(self, estimator_config: dict, loss_config: dict):
+        self.use_leave_one_out_baseline = estimator_config["use_leave_one_out_baseline"]
+        self.normalize_rewards = estimator_config["normalize_rewards"]
+
+    def compute_advantage(
+        self,
+        prompt_ids,
+        rewards,
+        repeated_batch,
+        mask,
+        **kwargs,
+    ):
+        """Compute GDPO advantages.
+
+        Args:
+            prompt_ids: Tensor identifying which prompt each sample belongs to (for per-prompt baselines).
+            rewards: Unused; for interface consistency.
+            repeated_batch: Batch containing reward1, reward2, ... keys.
+            mask: Response token mask of shape [batch_size, seq_len], 1 for valid response tokens, 0 for padding.
+            **kwargs: Additional arguments (unused).
+
+        Returns:
+            Advantages tensor of shape [batch_size, seq_len].
+        """
+        reward_component_keys = get_gdpo_reward_component_keys(repeated_batch)
+        if len(reward_component_keys) < 2:
+            raise ValueError(
+                f"GDPO requires multiple reward components (reward1, reward2, ...). "
+                f"This batch has {len(reward_component_keys)} component(s). "
+                "Switch to GRPO by setting grpo.adv_estimator.name to 'grpo' in your config."
+            )
+        valid = torch.ones_like(repeated_batch[reward_component_keys[0]])
+        leave_one_out = self.use_leave_one_out_baseline
+        assert prompt_ids.shape[0] == valid.shape[0], (
+            "prompt_ids must match reward batch size; "
+            f"got {prompt_ids.shape[0]} vs {valid.shape[0]}"
+        )
+        advantage_parts = []
+        for key in reward_component_keys:
+            r = repeated_batch[key]
+            base, std_k = calculate_baseline_and_std_per_prompt(
+                prompt_ids,
+                r,
+                valid,
+                leave_one_out_baseline=leave_one_out,
+            )
+            adv_k = (r - base).unsqueeze(-1)
+            if self.normalize_rewards:
+                epsilon = 1e-6
+                non_zero_std_mask = std_k > 0
+                adv_k[non_zero_std_mask] = adv_k[non_zero_std_mask] / (
+                    std_k.unsqueeze(-1)[non_zero_std_mask] + epsilon
+                )
+
+            advantage_parts.append(adv_k)
+
+        advantages = sum(advantage_parts)
+        # Normalize combined advantage to zero mean and unit std
+        adv_std = advantages.std()
+        if adv_std > 0:
+            advantages = (advantages - advantages.mean()) / adv_std
+        else:
+            advantages = advantages - advantages.mean()
+
+        return advantages.expand(mask.shape)
+
+
 class ReinforcePlusPlusAdvantageEstimator:
     """Reinforce++ advantage estimator with optional baseline subtraction and KL penalty in reward.
 
@@ -87,6 +172,7 @@ def compute_advantage(
         self,
         prompt_ids,
         rewards,
+        repeated_batch,
         mask,
         logprobs_policy=None,
         logprobs_reference=None,
@@ -95,13 +181,12 @@ def compute_advantage(
         """Compute Reinforce++ advantages with optional KL penalty.
 
         Args:
-            prompt_ids: Tensor of shape [batch_size] identifying which prompt each sample belongs to.
+            prompt_ids: Tensor identifying which prompt each sample belongs to (for baseline).
             rewards: Tensor of shape [batch_size] containing reward for each sample.
+            repeated_batch: Unused; for interface consistency.
             mask: Response token mask of shape [batch_size, seq_len], 1 for valid response tokens, 0 for padding.
-                  Used for: (1) expanding advantages to token-level shape, (2) global normalization
-                  that only considers valid tokens.
-            logprobs_policy: Policy log probabilities of shape [batch_size, seq_len], required if use_kl_in_reward.
-            logprobs_reference: Reference policy log probabilities of shape [batch_size, seq_len], required if use_kl_in_reward.
+            logprobs_policy: Policy log probabilities, required if use_kl_in_reward.
+            logprobs_reference: Reference policy log probabilities, required if use_kl_in_reward.
             **kwargs: Additional arguments (unused).
 
         Returns:

nemo_rl/algorithms/grpo.py

@@ -28,6 +28,7 @@
 from transformers.tokenization_utils_base import PreTrainedTokenizerBase
 
 from nemo_rl.algorithms.advantage_estimator import (
+    GDPOAdvantageEstimator,
     GRPOAdvantageEstimator,
     ReinforcePlusPlusAdvantageEstimator,
 )
@@ -43,6 +44,7 @@
 )
 from nemo_rl.algorithms.utils import (
     calculate_baseline_and_std_per_prompt,
+    get_gdpo_reward_component_keys,
     log_generation_metrics_to_wandb,
     print_performance_metrics,
     set_seed,
@@ -121,9 +123,9 @@ class AsyncGRPOConfig(TypedDict):
 
 
 class AdvEstimatorConfig(TypedDict):
-    """Configuration for advantage estimator (GRPO or Reinforce++)."""
+    """Configuration for advantage estimator (GRPO, GDPO, or Reinforce++)."""
 
-    name: str  # "grpo" or "reinforce_plus_plus"
+    name: str  # "grpo", "gdpo", or "reinforce_plus_plus"
     # GRPO specific
     normalize_rewards: NotRequired[bool]
     use_leave_one_out_baseline: NotRequired[bool]
@@ -966,11 +968,16 @@ def scale_rewards(
             )
 
         # Clamp and scale
-        rewards = torch.clamp(rewards, min=source_min, max=source_max)
-        scaled_rewards = target_min + (rewards - source_min) / (
-            source_max - source_min
-        ) * (target_max - target_min)
+        def _scale(reward_tensor: torch.Tensor) -> torch.Tensor:
+            r = torch.clamp(reward_tensor, min=source_min, max=source_max)
+            return target_min + (r - source_min) / (source_max - source_min) * (
+                target_max - target_min
+            )
+
+        scaled_rewards = _scale(rewards)
         repeated_batch["total_reward"] = scaled_rewards
+        for key in get_gdpo_reward_component_keys(repeated_batch):
+            repeated_batch[key] = _scale(repeated_batch[key])
 
     return repeated_batch
 
@@ -1031,7 +1038,7 @@ def _create_advantage_estimator(master_config: MasterConfig):
         master_config: The master configuration dictionary.
 
     Returns:
-        An advantage estimator instance (GRPOAdvantageEstimator or ReinforcePlusPlusAdvantageEstimator).
+        An advantage estimator instance (GRPO, GDPO, or ReinforcePlusPlus).
 
     Raises:
         ValueError: If the advantage estimator name is not recognized.
@@ -1055,7 +1062,14 @@ def _create_advantage_estimator(master_config: MasterConfig):
     )
 
     adv_estimator_name = adv_estimator_config["name"]
-    if adv_estimator_name == "grpo":
+    if adv_estimator_name == "gdpo":
+        assert not _should_use_async_rollouts(master_config), (
+            "GDPO is not supported for async rollouts, "
+            "please set policy.generation.vllm_cfg.async_engine to false in your config."
+        )
+        adv_estimator = GDPOAdvantageEstimator(adv_estimator_config, loss_config)
+        print("  ✓ Using GDPO advantage estimator (multi-reward)")
+    elif adv_estimator_name == "grpo":
         adv_estimator = GRPOAdvantageEstimator(adv_estimator_config, loss_config)
         print("  ✓ Using GRPO advantage estimator")
     elif adv_estimator_name == "reinforce_plus_plus":
@@ -1644,10 +1658,10 @@ def grpo_train(
                     # If the current batch is not enough to fill the buffer during dynamic sampling, we update the cache and process the next batch.
                     if not is_batch_complete:
                         continue
+
                     gen_step_metrics = {}
                     if hasattr(policy_generation, "get_step_metrics"):
                         gen_step_metrics = policy_generation.get_step_metrics()
-                    advantages = (rewards - baseline).unsqueeze(-1)
 
                     # Save baseline for logging (before deletion)
                     baseline_for_log = baseline.clone()
@@ -1778,6 +1792,7 @@ def grpo_train(
                     train_data["advantages"] = adv_estimator.compute_advantage(
                         prompt_ids=prompt_ids_for_adv,
                         rewards=rewards,
+                        repeated_batch=repeated_batch,
                         mask=mask,
                         logprobs_policy=train_data["prev_logprobs"],
                         logprobs_reference=train_data.get("reference_policy_logprobs"),
@@ -2809,6 +2824,7 @@ def async_grpo_train(
                     train_data["advantages"] = adv_estimator.compute_advantage(
                         prompt_ids=prompt_ids_for_adv,
                         rewards=rewards,
+                        repeated_batch=repeated_batch,
                         mask=mask,
                         logprobs_policy=train_data["prev_logprobs"],
                         logprobs_reference=train_data.get("reference_policy_logprobs"),

nemo_rl/algorithms/utils.py

@@ -14,6 +14,7 @@
 
 import math
 import random
+import re
 import warnings
 from functools import partial, wraps
 from typing import Any, Optional
@@ -31,6 +32,12 @@
 from nemo_rl.utils.logger import Logger
 
 
+def get_gdpo_reward_component_keys(batch) -> list:
+    """Return batch keys that are reward components (reward1, reward2, ...) in sorted order."""
+    keys = [k for k in batch.keys() if re.match(r"reward\d+$", str(k))]
+    return sorted(keys, key=lambda k: int(re.search(r"\d+", str(k)).group()))
+
+
 def calculate_kl(
     logprobs: torch.Tensor,
     logprobs_reference: torch.Tensor,

nemo_rl/data/datasets/response_datasets/__init__.py

@@ -25,6 +25,7 @@
     GeneralConversationsJsonlDataset,
 )
 from nemo_rl.data.datasets.response_datasets.geometry3k import Geometry3KDataset
+from nemo_rl.data.datasets.response_datasets.gsm8k import GSM8KDataset
 from nemo_rl.data.datasets.response_datasets.helpsteer3 import HelpSteer3Dataset
 from nemo_rl.data.datasets.response_datasets.nemogym_dataset import NemoGymDataset
 from nemo_rl.data.datasets.response_datasets.oai_format_dataset import (
@@ -55,6 +56,7 @@
     "refcoco": RefCOCODataset,
     "squad": SquadDataset,
     "tulu3_sft_mixture": Tulu3SftMixtureDataset,
+    "gsm8k": GSM8KDataset,
     # load from local JSONL file or HuggingFace
     "openai_format": OpenAIFormatDataset,
     "NemoGymDataset": NemoGymDataset,
@@ -94,6 +96,7 @@ def load_response_dataset(data_config: ResponseDatasetConfig):
     "GeneralConversationsJsonlDataset",
     "DAPOMath17KDataset",
     "DAPOMathAIME2024Dataset",
+    "GSM8KDataset",
     "DeepScalerDataset",
     "Geometry3KDataset",
     "HelpSteer3Dataset",